| 5 |
5 |
// Author: Simon-Pierre Boucher |
| 6 |
6 |
// Mail: contact@spboucher.ai |
| 7 |
7 |
// |
| 8 |
|
−// Built-in model catalog, generated from docs/PROVIDERS.md (2026-07-30). |
| 9 |
|
−// PLACEHOLDER seed — replaced by the full generated catalog before Phase 6. |
| 10 |
|
−// When research and reality disagree in Phase 7, update docs/PROVIDERS.md and |
| 11 |
|
−// this file together. |
|
8 |
+// Built-in model catalog, generated from docs/PROVIDERS.md and the per-provider |
|
9 |
+// research files in docs/research/ on 2026-07-30. This file and docs/PROVIDERS.md |
|
10 |
+// are a single source of truth and MUST stay in sync: when Phase 7 verification |
|
11 |
+// (or any later research pass) changes a model, update both together. |
|
12 |
+// |
|
13 |
+// Scope: chat-completions-capable chat models only. Embeddings, audio/TTS/ASR, |
|
14 |
+// realtime, image/video generation, moderation, OCR, robotics, deep-research |
|
15 |
+// agents, and Responses-API-only models are excluded by design. |
| 12 |
16 |
// |
| 13 |
17 |
|
| 14 |
18 |
import Foundation |
| 15 |
19 |
|
| 16 |
20 |
enum ModelCatalogData { |
| 17 |
|
− static let all: [AIModel] = [ |
|
21 |
+ |
|
22 |
+ // MARK: - Shared parameter-support presets |
|
23 |
+ |
|
24 |
+ /// OpenAI reasoning models (o-series, gpt-5.x reasoning variants): reject |
|
25 |
+ /// temperature/top_p/penalties, require max_completion_tokens, accept reasoning_effort. |
|
26 |
+ private static let openAIReasoning = ParameterSupport( |
|
27 |
+ temperature: false, topP: false, |
|
28 |
+ usesMaxCompletionTokens: true, reasoningEffort: true |
|
29 |
+ ) |
|
30 |
+ |
|
31 |
+ // MARK: - OpenAI |
|
32 |
+ |
|
33 |
+ static let openai: [AIModel] = [ |
|
34 |
+ // Flagship GPT-5.6 trio |
|
35 |
+ AIModel( |
|
36 |
+ id: "gpt-5.6-sol", provider: .openai, displayName: "GPT-5.6 Sol", |
|
37 |
+ contextWindow: 1_050_000, maxOutputTokens: 128_000, |
|
38 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
39 |
+ pricing: ModelPricing(inputPerMTok: 5.00, outputPerMTok: 30.00), |
|
40 |
+ parameterSupport: openAIReasoning, |
|
41 |
+ isRecommended: true |
|
42 |
+ ), |
|
43 |
+ AIModel( |
|
44 |
+ id: "gpt-5.6-terra", provider: .openai, displayName: "GPT-5.6 Terra", |
|
45 |
+ contextWindow: 1_050_000, maxOutputTokens: 128_000, |
|
46 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
47 |
+ pricing: ModelPricing(inputPerMTok: 2.50, outputPerMTok: 15.00), |
|
48 |
+ parameterSupport: openAIReasoning, |
|
49 |
+ isRecommended: true |
|
50 |
+ ), |
|
51 |
+ AIModel( |
|
52 |
+ id: "gpt-5.6-luna", provider: .openai, displayName: "GPT-5.6 Luna", |
|
53 |
+ contextWindow: 1_050_000, maxOutputTokens: 128_000, |
|
54 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
55 |
+ pricing: ModelPricing(inputPerMTok: 1.00, outputPerMTok: 6.00), |
|
56 |
+ parameterSupport: openAIReasoning |
|
57 |
+ ), |
|
58 |
+ AIModel( |
|
59 |
+ id: "chat-latest", provider: .openai, displayName: "ChatGPT Latest", |
|
60 |
+ contextWindow: 128_000, maxOutputTokens: nil, |
|
61 |
+ capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
|
62 |
+ pricing: ModelPricing(inputPerMTok: 5.00, outputPerMTok: 30.00), |
|
63 |
+ parameterSupport: .openAIDefault |
|
64 |
+ ), |
|
65 |
+ // Current / recent GPT-5.x |
|
66 |
+ AIModel( |
|
67 |
+ id: "gpt-5.5", provider: .openai, displayName: "GPT-5.5", |
|
68 |
+ contextWindow: 400_000, maxOutputTokens: nil, |
|
69 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
70 |
+ pricing: ModelPricing(inputPerMTok: 5.00, outputPerMTok: 30.00), |
|
71 |
+ parameterSupport: openAIReasoning |
|
72 |
+ ), |
|
73 |
+ AIModel( |
|
74 |
+ id: "gpt-5.4", provider: .openai, displayName: "GPT-5.4", |
|
75 |
+ contextWindow: 400_000, maxOutputTokens: 128_000, |
|
76 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
77 |
+ pricing: ModelPricing(inputPerMTok: 2.50, outputPerMTok: 15.00), |
|
78 |
+ parameterSupport: openAIReasoning |
|
79 |
+ ), |
|
80 |
+ AIModel( |
|
81 |
+ id: "gpt-5.4-mini", provider: .openai, displayName: "GPT-5.4 mini", |
|
82 |
+ contextWindow: 400_000, maxOutputTokens: nil, |
|
83 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
84 |
+ pricing: ModelPricing(inputPerMTok: 0.75, outputPerMTok: 4.50), |
|
85 |
+ parameterSupport: openAIReasoning |
|
86 |
+ ), |
|
87 |
+ AIModel( |
|
88 |
+ id: "gpt-5.4-nano", provider: .openai, displayName: "GPT-5.4 nano", |
|
89 |
+ contextWindow: 400_000, maxOutputTokens: nil, |
|
90 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
91 |
+ pricing: ModelPricing(inputPerMTok: 0.20, outputPerMTok: 1.25), |
|
92 |
+ parameterSupport: openAIReasoning |
|
93 |
+ ), |
|
94 |
+ AIModel( |
|
95 |
+ id: "gpt-5.3-chat-latest", provider: .openai, displayName: "GPT-5.3 Chat Latest", |
|
96 |
+ contextWindow: 128_000, maxOutputTokens: nil, |
|
97 |
+ capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
|
98 |
+ pricing: nil, |
|
99 |
+ parameterSupport: .openAIDefault |
|
100 |
+ ), |
|
101 |
+ AIModel( |
|
102 |
+ id: "gpt-5.2", provider: .openai, displayName: "GPT-5.2", |
|
103 |
+ contextWindow: 400_000, maxOutputTokens: 128_000, |
|
104 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
105 |
+ pricing: ModelPricing(inputPerMTok: 1.75, outputPerMTok: 14.00), |
|
106 |
+ parameterSupport: openAIReasoning |
|
107 |
+ ), |
|
108 |
+ AIModel( |
|
109 |
+ id: "gpt-5.2-chat-latest", provider: .openai, displayName: "GPT-5.2 Chat Latest", |
|
110 |
+ contextWindow: 128_000, maxOutputTokens: 16_000, |
|
111 |
+ capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
|
112 |
+ pricing: ModelPricing(inputPerMTok: 1.75, outputPerMTok: 14.00), |
|
113 |
+ parameterSupport: .openAIDefault |
|
114 |
+ ), |
|
115 |
+ AIModel( |
|
116 |
+ id: "gpt-5.1", provider: .openai, displayName: "GPT-5.1", |
|
117 |
+ contextWindow: 400_000, maxOutputTokens: 128_000, |
|
118 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
119 |
+ pricing: ModelPricing(inputPerMTok: 1.25, outputPerMTok: 10.00), |
|
120 |
+ parameterSupport: openAIReasoning |
|
121 |
+ ), |
|
122 |
+ AIModel( |
|
123 |
+ id: "gpt-5.1-chat-latest", provider: .openai, displayName: "GPT-5.1 Chat Latest", |
|
124 |
+ contextWindow: 128_000, maxOutputTokens: 16_000, |
|
125 |
+ capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
|
126 |
+ pricing: ModelPricing(inputPerMTok: 1.25, outputPerMTok: 10.00), |
|
127 |
+ parameterSupport: .openAIDefault |
|
128 |
+ ), |
|
129 |
+ AIModel( |
|
130 |
+ id: "gpt-5", provider: .openai, displayName: "GPT-5", |
|
131 |
+ contextWindow: 400_000, maxOutputTokens: 128_000, |
|
132 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
133 |
+ pricing: ModelPricing(inputPerMTok: 1.25, outputPerMTok: 10.00), |
|
134 |
+ parameterSupport: openAIReasoning |
|
135 |
+ ), |
|
136 |
+ AIModel( |
|
137 |
+ id: "gpt-5-mini", provider: .openai, displayName: "GPT-5 mini", |
|
138 |
+ contextWindow: 400_000, maxOutputTokens: 128_000, |
|
139 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
140 |
+ pricing: ModelPricing(inputPerMTok: 0.25, outputPerMTok: 2.00), |
|
141 |
+ parameterSupport: openAIReasoning |
|
142 |
+ ), |
|
143 |
+ AIModel( |
|
144 |
+ id: "gpt-5-nano", provider: .openai, displayName: "GPT-5 nano", |
|
145 |
+ contextWindow: 400_000, maxOutputTokens: 128_000, |
|
146 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
147 |
+ pricing: ModelPricing(inputPerMTok: 0.05, outputPerMTok: 0.40), |
|
148 |
+ parameterSupport: openAIReasoning |
|
149 |
+ ), |
|
150 |
+ AIModel( |
|
151 |
+ id: "gpt-5-chat-latest", provider: .openai, displayName: "GPT-5 Chat Latest", |
|
152 |
+ contextWindow: 128_000, maxOutputTokens: 16_000, |
|
153 |
+ capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
|
154 |
+ pricing: ModelPricing(inputPerMTok: 1.25, outputPerMTok: 10.00), |
|
155 |
+ parameterSupport: .openAIDefault |
|
156 |
+ ), |
|
157 |
+ // o-series reasoning |
|
158 |
+ AIModel( |
|
159 |
+ id: "o3", provider: .openai, displayName: "OpenAI o3", |
|
160 |
+ contextWindow: 200_000, maxOutputTokens: 100_000, |
|
161 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
162 |
+ pricing: ModelPricing(inputPerMTok: 2.00, outputPerMTok: 8.00), |
|
163 |
+ parameterSupport: openAIReasoning |
|
164 |
+ ), |
|
165 |
+ AIModel( |
|
166 |
+ id: "o4-mini", provider: .openai, displayName: "OpenAI o4-mini", |
|
167 |
+ contextWindow: 200_000, maxOutputTokens: 100_000, |
|
168 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
169 |
+ pricing: ModelPricing(inputPerMTok: 1.10, outputPerMTok: 4.40), |
|
170 |
+ parameterSupport: openAIReasoning |
|
171 |
+ ), |
| 18 |
172 |
AIModel( |
| 19 |
|
− id: "gpt-4o-mini", |
| 20 |
|
− provider: .openai, |
| 21 |
|
− displayName: "GPT-4o mini", |
| 22 |
|
− contextWindow: 128_000, |
| 23 |
|
− maxOutputTokens: 16_384, |
|
173 |
+ id: "o3-mini", provider: .openai, displayName: "OpenAI o3-mini", |
|
174 |
+ contextWindow: 200_000, maxOutputTokens: 100_000, |
|
175 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
176 |
+ pricing: ModelPricing(inputPerMTok: 1.10, outputPerMTok: 4.40), |
|
177 |
+ parameterSupport: openAIReasoning, |
|
178 |
+ isLegacy: true |
|
179 |
+ ), |
|
180 |
+ AIModel( |
|
181 |
+ id: "o1", provider: .openai, displayName: "OpenAI o1", |
|
182 |
+ contextWindow: 200_000, maxOutputTokens: 100_000, |
|
183 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
184 |
+ pricing: ModelPricing(inputPerMTok: 15.00, outputPerMTok: 60.00), |
|
185 |
+ parameterSupport: openAIReasoning, |
|
186 |
+ isLegacy: true |
|
187 |
+ ), |
|
188 |
+ // Legacy GPT-4.x / 3.5 |
|
189 |
+ AIModel( |
|
190 |
+ id: "gpt-4.1", provider: .openai, displayName: "GPT-4.1", |
|
191 |
+ contextWindow: 1_047_576, maxOutputTokens: 32_768, |
|
192 |
+ capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
|
193 |
+ pricing: ModelPricing(inputPerMTok: 2.00, outputPerMTok: 8.00), |
|
194 |
+ parameterSupport: .openAIDefault, |
|
195 |
+ isLegacy: true |
|
196 |
+ ), |
|
197 |
+ AIModel( |
|
198 |
+ id: "gpt-4.1-mini", provider: .openai, displayName: "GPT-4.1 mini", |
|
199 |
+ contextWindow: 1_047_576, maxOutputTokens: 32_768, |
|
200 |
+ capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
|
201 |
+ pricing: ModelPricing(inputPerMTok: 0.40, outputPerMTok: 1.60), |
|
202 |
+ parameterSupport: .openAIDefault, |
|
203 |
+ isLegacy: true |
|
204 |
+ ), |
|
205 |
+ AIModel( |
|
206 |
+ id: "gpt-4.1-nano", provider: .openai, displayName: "GPT-4.1 nano", |
|
207 |
+ contextWindow: 1_047_576, maxOutputTokens: 32_768, |
|
208 |
+ capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
|
209 |
+ pricing: ModelPricing(inputPerMTok: 0.10, outputPerMTok: 0.40), |
|
210 |
+ parameterSupport: .openAIDefault, |
|
211 |
+ isLegacy: true |
|
212 |
+ ), |
|
213 |
+ AIModel( |
|
214 |
+ id: "gpt-4o", provider: .openai, displayName: "GPT-4o", |
|
215 |
+ contextWindow: 128_000, maxOutputTokens: 16_384, |
|
216 |
+ capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
|
217 |
+ pricing: ModelPricing(inputPerMTok: 2.50, outputPerMTok: 10.00), |
|
218 |
+ parameterSupport: .openAIDefault, |
|
219 |
+ isLegacy: true |
|
220 |
+ ), |
|
221 |
+ AIModel( |
|
222 |
+ id: "gpt-4o-mini", provider: .openai, displayName: "GPT-4o mini", |
|
223 |
+ contextWindow: 128_000, maxOutputTokens: 16_384, |
| 24 |
224 |
capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
| 25 |
225 |
pricing: ModelPricing(inputPerMTok: 0.15, outputPerMTok: 0.60), |
| 26 |
226 |
parameterSupport: .openAIDefault, |
|
227 |
+ isLegacy: true |
|
228 |
+ ), |
|
229 |
+ AIModel( |
|
230 |
+ id: "gpt-4-turbo", provider: .openai, displayName: "GPT-4 Turbo", |
|
231 |
+ contextWindow: 128_000, maxOutputTokens: 4_096, |
|
232 |
+ capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
|
233 |
+ pricing: ModelPricing(inputPerMTok: 10.00, outputPerMTok: 30.00), |
|
234 |
+ parameterSupport: .openAIDefault, |
|
235 |
+ isLegacy: true |
|
236 |
+ ), |
|
237 |
+ AIModel( |
|
238 |
+ id: "gpt-4", provider: .openai, displayName: "GPT-4", |
|
239 |
+ contextWindow: 8_192, maxOutputTokens: 8_192, |
|
240 |
+ capabilities: ModelCapabilities(tools: true), |
|
241 |
+ pricing: ModelPricing(inputPerMTok: 30.00, outputPerMTok: 60.00), |
|
242 |
+ parameterSupport: .openAIDefault, |
|
243 |
+ isLegacy: true |
|
244 |
+ ), |
|
245 |
+ AIModel( |
|
246 |
+ id: "gpt-3.5-turbo", provider: .openai, displayName: "GPT-3.5 Turbo", |
|
247 |
+ contextWindow: 16_385, maxOutputTokens: 4_096, |
|
248 |
+ capabilities: ModelCapabilities(tools: true, jsonMode: true), |
|
249 |
+ pricing: ModelPricing(inputPerMTok: 0.50, outputPerMTok: 1.50), |
|
250 |
+ parameterSupport: .openAIDefault, |
|
251 |
+ isLegacy: true |
|
252 |
+ ), |
|
253 |
+ ] |
|
254 |
+ |
|
255 |
+ // MARK: - Anthropic |
|
256 |
+ |
|
257 |
+ static let anthropic: [AIModel] = [ |
|
258 |
+ AIModel( |
|
259 |
+ id: "claude-opus-5", provider: .anthropic, displayName: "Claude Opus 5", |
|
260 |
+ contextWindow: 1_000_000, maxOutputTokens: 128_000, |
|
261 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
262 |
+ pricing: ModelPricing(inputPerMTok: 5.00, outputPerMTok: 25.00), |
|
263 |
+ parameterSupport: ParameterSupport(temperature: false, topP: false, thinkingToggle: true), |
| 27 |
264 |
isRecommended: true |
| 28 |
265 |
), |
| 29 |
266 |
AIModel( |
| 30 |
|
− id: "claude-sonnet-5", |
| 31 |
|
− provider: .anthropic, |
| 32 |
|
− displayName: "Claude Sonnet 5", |
| 33 |
|
− contextWindow: 1_000_000, |
| 34 |
|
− maxOutputTokens: 128_000, |
|
267 |
+ id: "claude-sonnet-5", provider: .anthropic, displayName: "Claude Sonnet 5", |
|
268 |
+ contextWindow: 1_000_000, maxOutputTokens: 128_000, |
| 35 |
269 |
capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
| 36 |
|
− pricing: ModelPricing(inputPerMTok: 2, outputPerMTok: 10), |
|
270 |
+ pricing: ModelPricing(inputPerMTok: 3.00, outputPerMTok: 15.00), |
| 37 |
271 |
parameterSupport: ParameterSupport(temperature: false, topP: false, thinkingToggle: true), |
| 38 |
272 |
isRecommended: true |
| 39 |
273 |
), |
|
274 |
+ AIModel( |
|
275 |
+ id: "claude-fable-5", provider: .anthropic, displayName: "Claude Fable 5", |
|
276 |
+ contextWindow: 1_000_000, maxOutputTokens: 128_000, |
|
277 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
278 |
+ pricing: ModelPricing(inputPerMTok: 10.00, outputPerMTok: 50.00), |
|
279 |
+ // Thinking is always on and cannot be disabled — no toggle. |
|
280 |
+ parameterSupport: ParameterSupport(temperature: false, topP: false, thinkingToggle: false) |
|
281 |
+ ), |
|
282 |
+ AIModel( |
|
283 |
+ id: "claude-opus-4-8", provider: .anthropic, displayName: "Claude Opus 4.8", |
|
284 |
+ contextWindow: 1_000_000, maxOutputTokens: 128_000, |
|
285 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
286 |
+ pricing: ModelPricing(inputPerMTok: 5.00, outputPerMTok: 25.00), |
|
287 |
+ parameterSupport: ParameterSupport(temperature: false, topP: false, thinkingToggle: true) |
|
288 |
+ ), |
|
289 |
+ AIModel( |
|
290 |
+ id: "claude-opus-4-7", provider: .anthropic, displayName: "Claude Opus 4.7", |
|
291 |
+ contextWindow: 1_000_000, maxOutputTokens: 128_000, |
|
292 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
293 |
+ pricing: ModelPricing(inputPerMTok: 5.00, outputPerMTok: 25.00), |
|
294 |
+ parameterSupport: ParameterSupport(temperature: false, topP: false, thinkingToggle: true) |
|
295 |
+ ), |
|
296 |
+ AIModel( |
|
297 |
+ id: "claude-opus-4-6", provider: .anthropic, displayName: "Claude Opus 4.6", |
|
298 |
+ contextWindow: 1_000_000, maxOutputTokens: 128_000, |
|
299 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
300 |
+ pricing: ModelPricing(inputPerMTok: 5.00, outputPerMTok: 25.00), |
|
301 |
+ parameterSupport: ParameterSupport(temperature: true, topP: true, thinkingToggle: true) |
|
302 |
+ ), |
|
303 |
+ AIModel( |
|
304 |
+ id: "claude-sonnet-4-6", provider: .anthropic, displayName: "Claude Sonnet 4.6", |
|
305 |
+ contextWindow: 1_000_000, maxOutputTokens: 128_000, |
|
306 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
307 |
+ pricing: ModelPricing(inputPerMTok: 3.00, outputPerMTok: 15.00), |
|
308 |
+ parameterSupport: ParameterSupport(temperature: true, topP: true, thinkingToggle: true) |
|
309 |
+ ), |
|
310 |
+ AIModel( |
|
311 |
+ id: "claude-haiku-4-5-20251001", provider: .anthropic, displayName: "Claude Haiku 4.5", |
|
312 |
+ contextWindow: 200_000, maxOutputTokens: 64_000, |
|
313 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
314 |
+ pricing: ModelPricing(inputPerMTok: 1.00, outputPerMTok: 5.00), |
|
315 |
+ parameterSupport: ParameterSupport(temperature: true, topP: true, thinkingToggle: true) |
|
316 |
+ ), |
|
317 |
+ AIModel( |
|
318 |
+ id: "claude-opus-4-5-20251101", provider: .anthropic, displayName: "Claude Opus 4.5", |
|
319 |
+ contextWindow: 200_000, maxOutputTokens: 64_000, |
|
320 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
321 |
+ pricing: ModelPricing(inputPerMTok: 5.00, outputPerMTok: 25.00), |
|
322 |
+ parameterSupport: ParameterSupport(temperature: true, topP: true, thinkingToggle: true), |
|
323 |
+ isLegacy: true |
|
324 |
+ ), |
|
325 |
+ AIModel( |
|
326 |
+ id: "claude-sonnet-4-5-20250929", provider: .anthropic, displayName: "Claude Sonnet 4.5", |
|
327 |
+ contextWindow: 1_000_000, maxOutputTokens: 64_000, |
|
328 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
329 |
+ pricing: ModelPricing(inputPerMTok: 3.00, outputPerMTok: 15.00), |
|
330 |
+ parameterSupport: ParameterSupport(temperature: true, topP: true, thinkingToggle: true), |
|
331 |
+ isLegacy: true |
|
332 |
+ ), |
|
333 |
+ AIModel( |
|
334 |
+ id: "claude-opus-4-1-20250805", provider: .anthropic, displayName: "Claude Opus 4.1", |
|
335 |
+ contextWindow: 200_000, maxOutputTokens: 32_000, |
|
336 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
337 |
+ pricing: ModelPricing(inputPerMTok: 15.00, outputPerMTok: 75.00), |
|
338 |
+ parameterSupport: ParameterSupport(temperature: true, topP: true, thinkingToggle: true), |
|
339 |
+ isLegacy: true |
|
340 |
+ ), |
|
341 |
+ ] |
|
342 |
+ |
|
343 |
+ // MARK: - xAI (Grok) |
|
344 |
+ |
|
345 |
+ static let xai: [AIModel] = [ |
|
346 |
+ AIModel( |
|
347 |
+ id: "grok-4.5", provider: .xai, displayName: "Grok 4.5", |
|
348 |
+ contextWindow: 500_000, maxOutputTokens: nil, |
|
349 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
350 |
+ pricing: ModelPricing(inputPerMTok: 2.00, outputPerMTok: 6.00), |
|
351 |
+ parameterSupport: ParameterSupport(reasoningEffort: true), |
|
352 |
+ isRecommended: true |
|
353 |
+ ), |
|
354 |
+ AIModel( |
|
355 |
+ id: "grok-4.3", provider: .xai, displayName: "Grok 4.3", |
|
356 |
+ contextWindow: 1_000_000, maxOutputTokens: nil, |
|
357 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
358 |
+ pricing: ModelPricing(inputPerMTok: 1.25, outputPerMTok: 2.50), |
|
359 |
+ parameterSupport: ParameterSupport(reasoningEffort: true) |
|
360 |
+ ), |
|
361 |
+ AIModel( |
|
362 |
+ id: "grok-4.20", provider: .xai, displayName: "Grok 4.20 Reasoning", |
|
363 |
+ contextWindow: 1_000_000, maxOutputTokens: nil, |
|
364 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
365 |
+ pricing: ModelPricing(inputPerMTok: 1.25, outputPerMTok: 2.50), |
|
366 |
+ parameterSupport: ParameterSupport(reasoningEffort: true) |
|
367 |
+ ), |
|
368 |
+ AIModel( |
|
369 |
+ id: "grok-4.20-non-reasoning", provider: .xai, displayName: "Grok 4.20 Non-Reasoning", |
|
370 |
+ contextWindow: 1_000_000, maxOutputTokens: nil, |
|
371 |
+ capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
|
372 |
+ pricing: ModelPricing(inputPerMTok: 1.25, outputPerMTok: 2.50), |
|
373 |
+ parameterSupport: ParameterSupport() |
|
374 |
+ ), |
|
375 |
+ AIModel( |
|
376 |
+ id: "grok-4.20-multi-agent", provider: .xai, displayName: "Grok 4.20 Multi-Agent", |
|
377 |
+ contextWindow: 1_000_000, maxOutputTokens: nil, |
|
378 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
379 |
+ pricing: ModelPricing(inputPerMTok: 1.25, outputPerMTok: 2.50), |
|
380 |
+ parameterSupport: ParameterSupport(reasoningEffort: true) |
|
381 |
+ ), |
|
382 |
+ AIModel( |
|
383 |
+ id: "grok-code-fast-1", provider: .xai, displayName: "Grok Code Fast 1", |
|
384 |
+ contextWindow: 256_000, maxOutputTokens: nil, |
|
385 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
386 |
+ pricing: ModelPricing(inputPerMTok: 1.00, outputPerMTok: 2.00), |
|
387 |
+ parameterSupport: ParameterSupport(reasoningEffort: true), |
|
388 |
+ isRecommended: true |
|
389 |
+ ), |
|
390 |
+ ] |
|
391 |
+ |
|
392 |
+ // MARK: - Mistral |
|
393 |
+ |
|
394 |
+ static let mistral: [AIModel] = [ |
|
395 |
+ AIModel( |
|
396 |
+ id: "mistral-medium-latest", provider: .mistral, displayName: "Mistral Medium 3.5", |
|
397 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
398 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
399 |
+ pricing: ModelPricing(inputPerMTok: 1.50, outputPerMTok: 7.50), |
|
400 |
+ parameterSupport: ParameterSupport(frequencyPenalty: true, presencePenalty: true, reasoningEffort: true), |
|
401 |
+ isRecommended: true |
|
402 |
+ ), |
|
403 |
+ AIModel( |
|
404 |
+ id: "mistral-large-latest", provider: .mistral, displayName: "Mistral Large 3", |
|
405 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
406 |
+ capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
|
407 |
+ pricing: ModelPricing(inputPerMTok: 0.50, outputPerMTok: 1.50), |
|
408 |
+ parameterSupport: .openAIDefault, |
|
409 |
+ isRecommended: true |
|
410 |
+ ), |
|
411 |
+ AIModel( |
|
412 |
+ id: "mistral-small-latest", provider: .mistral, displayName: "Mistral Small 4", |
|
413 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
414 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
415 |
+ pricing: ModelPricing(inputPerMTok: 0.15, outputPerMTok: 0.60), |
|
416 |
+ parameterSupport: ParameterSupport(frequencyPenalty: true, presencePenalty: true, reasoningEffort: true), |
|
417 |
+ isRecommended: true |
|
418 |
+ ), |
|
419 |
+ AIModel( |
|
420 |
+ id: "codestral-latest", provider: .mistral, displayName: "Codestral", |
|
421 |
+ contextWindow: 256_000, maxOutputTokens: nil, |
|
422 |
+ capabilities: ModelCapabilities(tools: true, jsonMode: true), |
|
423 |
+ pricing: ModelPricing(inputPerMTok: 0.30, outputPerMTok: 0.90), |
|
424 |
+ parameterSupport: .openAIDefault |
|
425 |
+ ), |
|
426 |
+ AIModel( |
|
427 |
+ id: "ministral-14b-latest", provider: .mistral, displayName: "Ministral 3 14B", |
|
428 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
429 |
+ capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
|
430 |
+ pricing: ModelPricing(inputPerMTok: 0.20, outputPerMTok: 0.20), |
|
431 |
+ parameterSupport: .openAIDefault |
|
432 |
+ ), |
|
433 |
+ AIModel( |
|
434 |
+ id: "ministral-8b-latest", provider: .mistral, displayName: "Ministral 3 8B", |
|
435 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
436 |
+ capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
|
437 |
+ pricing: ModelPricing(inputPerMTok: 0.15, outputPerMTok: 0.15), |
|
438 |
+ parameterSupport: .openAIDefault |
|
439 |
+ ), |
|
440 |
+ AIModel( |
|
441 |
+ id: "ministral-3b-latest", provider: .mistral, displayName: "Ministral 3 3B", |
|
442 |
+ contextWindow: 131_072, maxOutputTokens: nil, |
|
443 |
+ capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
|
444 |
+ pricing: ModelPricing(inputPerMTok: 0.10, outputPerMTok: 0.10), |
|
445 |
+ parameterSupport: .openAIDefault |
|
446 |
+ ), |
|
447 |
+ // Legacy / deprecated (still served; hidden by default) |
|
448 |
+ AIModel( |
|
449 |
+ id: "magistral-medium-latest", provider: .mistral, displayName: "Magistral Medium", |
|
450 |
+ contextWindow: 131_072, maxOutputTokens: nil, |
|
451 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
452 |
+ pricing: ModelPricing(inputPerMTok: 2.00, outputPerMTok: 5.00), |
|
453 |
+ parameterSupport: .openAIDefault, |
|
454 |
+ isLegacy: true |
|
455 |
+ ), |
|
456 |
+ AIModel( |
|
457 |
+ id: "devstral-latest", provider: .mistral, displayName: "Devstral 2", |
|
458 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
459 |
+ capabilities: ModelCapabilities(tools: true, jsonMode: true), |
|
460 |
+ pricing: ModelPricing(inputPerMTok: 0.40, outputPerMTok: 2.00), |
|
461 |
+ parameterSupport: .openAIDefault, |
|
462 |
+ isLegacy: true |
|
463 |
+ ), |
|
464 |
+ AIModel( |
|
465 |
+ id: "open-mistral-nemo", provider: .mistral, displayName: "Mistral Nemo", |
|
466 |
+ contextWindow: 131_072, maxOutputTokens: nil, |
|
467 |
+ capabilities: ModelCapabilities(tools: true, jsonMode: true), |
|
468 |
+ pricing: ModelPricing(inputPerMTok: 0.15, outputPerMTok: 0.15), |
|
469 |
+ parameterSupport: .openAIDefault, |
|
470 |
+ isLegacy: true |
|
471 |
+ ), |
|
472 |
+ ] |
|
473 |
+ |
|
474 |
+ // MARK: - Google Gemini |
|
475 |
+ |
|
476 |
+ static let gemini: [AIModel] = [ |
|
477 |
+ AIModel( |
|
478 |
+ id: "gemini-3.6-flash", provider: .gemini, displayName: "Gemini 3.6 Flash", |
|
479 |
+ contextWindow: 1_048_576, maxOutputTokens: 65_536, |
|
480 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
481 |
+ pricing: ModelPricing(inputPerMTok: 1.50, outputPerMTok: 7.50), |
|
482 |
+ parameterSupport: ParameterSupport(reasoningEffort: true), |
|
483 |
+ isRecommended: true |
|
484 |
+ ), |
|
485 |
+ AIModel( |
|
486 |
+ id: "gemini-3.5-flash", provider: .gemini, displayName: "Gemini 3.5 Flash", |
|
487 |
+ contextWindow: 1_048_576, maxOutputTokens: 65_536, |
|
488 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
489 |
+ pricing: ModelPricing(inputPerMTok: 1.50, outputPerMTok: 9.00), |
|
490 |
+ parameterSupport: ParameterSupport(reasoningEffort: true) |
|
491 |
+ ), |
|
492 |
+ AIModel( |
|
493 |
+ id: "gemini-3.5-flash-lite", provider: .gemini, displayName: "Gemini 3.5 Flash-Lite", |
|
494 |
+ contextWindow: 1_048_576, maxOutputTokens: 65_536, |
|
495 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
496 |
+ pricing: ModelPricing(inputPerMTok: 0.30, outputPerMTok: 2.50), |
|
497 |
+ parameterSupport: ParameterSupport(reasoningEffort: true), |
|
498 |
+ isRecommended: true |
|
499 |
+ ), |
|
500 |
+ AIModel( |
|
501 |
+ id: "gemini-3.1-pro-preview", provider: .gemini, displayName: "Gemini 3.1 Pro (Preview)", |
|
502 |
+ contextWindow: 1_048_576, maxOutputTokens: 65_536, |
|
503 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
504 |
+ pricing: ModelPricing(inputPerMTok: 2.00, outputPerMTok: 12.00), |
|
505 |
+ parameterSupport: ParameterSupport(reasoningEffort: true), |
|
506 |
+ isRecommended: true |
|
507 |
+ ), |
|
508 |
+ AIModel( |
|
509 |
+ id: "gemini-3.1-flash-lite", provider: .gemini, displayName: "Gemini 3.1 Flash-Lite", |
|
510 |
+ contextWindow: 1_048_576, maxOutputTokens: 65_536, |
|
511 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
512 |
+ pricing: ModelPricing(inputPerMTok: 0.25, outputPerMTok: 1.50), |
|
513 |
+ parameterSupport: ParameterSupport(reasoningEffort: true) |
|
514 |
+ ), |
|
515 |
+ AIModel( |
|
516 |
+ id: "gemini-2.5-pro", provider: .gemini, displayName: "Gemini 2.5 Pro", |
|
517 |
+ contextWindow: 1_048_576, maxOutputTokens: 65_536, |
|
518 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
519 |
+ pricing: ModelPricing(inputPerMTok: 1.25, outputPerMTok: 10.00), |
|
520 |
+ parameterSupport: ParameterSupport(reasoningEffort: true) |
|
521 |
+ ), |
|
522 |
+ AIModel( |
|
523 |
+ id: "gemini-2.5-flash", provider: .gemini, displayName: "Gemini 2.5 Flash", |
|
524 |
+ contextWindow: 1_048_576, maxOutputTokens: 65_536, |
|
525 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
526 |
+ pricing: ModelPricing(inputPerMTok: 0.30, outputPerMTok: 2.50), |
|
527 |
+ parameterSupport: ParameterSupport(reasoningEffort: true) |
|
528 |
+ ), |
|
529 |
+ AIModel( |
|
530 |
+ id: "gemini-2.5-flash-lite", provider: .gemini, displayName: "Gemini 2.5 Flash-Lite", |
|
531 |
+ contextWindow: 1_048_576, maxOutputTokens: 65_536, |
|
532 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
533 |
+ pricing: ModelPricing(inputPerMTok: 0.10, outputPerMTok: 0.40), |
|
534 |
+ parameterSupport: ParameterSupport(reasoningEffort: true) |
|
535 |
+ ), |
|
536 |
+ // Rolling aliases (auto-track the latest release; pricing varies with target) |
|
537 |
+ AIModel( |
|
538 |
+ id: "gemini-pro-latest", provider: .gemini, displayName: "Gemini Pro (Latest)", |
|
539 |
+ contextWindow: 1_048_576, maxOutputTokens: 65_536, |
|
540 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
541 |
+ pricing: nil, |
|
542 |
+ parameterSupport: ParameterSupport(reasoningEffort: true) |
|
543 |
+ ), |
|
544 |
+ AIModel( |
|
545 |
+ id: "gemini-flash-latest", provider: .gemini, displayName: "Gemini Flash (Latest)", |
|
546 |
+ contextWindow: 1_048_576, maxOutputTokens: 65_536, |
|
547 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
548 |
+ pricing: nil, |
|
549 |
+ parameterSupport: ParameterSupport(reasoningEffort: true) |
|
550 |
+ ), |
|
551 |
+ AIModel( |
|
552 |
+ id: "gemini-flash-lite-latest", provider: .gemini, displayName: "Gemini Flash-Lite (Latest)", |
|
553 |
+ contextWindow: 1_048_576, maxOutputTokens: 65_536, |
|
554 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
555 |
+ pricing: nil, |
|
556 |
+ parameterSupport: ParameterSupport(reasoningEffort: true) |
|
557 |
+ ), |
|
558 |
+ // Preview / secondary |
|
559 |
+ AIModel( |
|
560 |
+ id: "gemini-3-pro-preview", provider: .gemini, displayName: "Gemini 3 Pro (Preview)", |
|
561 |
+ contextWindow: 1_048_576, maxOutputTokens: 65_536, |
|
562 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
563 |
+ pricing: ModelPricing(inputPerMTok: 2.00, outputPerMTok: 12.00), |
|
564 |
+ parameterSupport: ParameterSupport(reasoningEffort: true) |
|
565 |
+ ), |
|
566 |
+ AIModel( |
|
567 |
+ id: "gemini-3-flash-preview", provider: .gemini, displayName: "Gemini 3 Flash (Preview)", |
|
568 |
+ contextWindow: 1_048_576, maxOutputTokens: 65_536, |
|
569 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
570 |
+ pricing: ModelPricing(inputPerMTok: 0.50, outputPerMTok: 3.00), |
|
571 |
+ parameterSupport: ParameterSupport(reasoningEffort: true) |
|
572 |
+ ), |
|
573 |
+ AIModel( |
|
574 |
+ id: "gemini-2.0-flash", provider: .gemini, displayName: "Gemini 2.0 Flash", |
|
575 |
+ contextWindow: 1_048_576, maxOutputTokens: 8_192, |
|
576 |
+ capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
|
577 |
+ pricing: ModelPricing(inputPerMTok: 0.10, outputPerMTok: 0.40), |
|
578 |
+ parameterSupport: ParameterSupport(), |
|
579 |
+ isLegacy: true |
|
580 |
+ ), |
|
581 |
+ AIModel( |
|
582 |
+ id: "gemini-2.0-flash-lite", provider: .gemini, displayName: "Gemini 2.0 Flash-Lite", |
|
583 |
+ contextWindow: 1_048_576, maxOutputTokens: 8_192, |
|
584 |
+ capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
|
585 |
+ pricing: ModelPricing(inputPerMTok: 0.075, outputPerMTok: 0.30), |
|
586 |
+ parameterSupport: ParameterSupport(), |
|
587 |
+ isLegacy: true |
|
588 |
+ ), |
|
589 |
+ AIModel( |
|
590 |
+ id: "gemma-4-26b-a4b-it", provider: .gemini, displayName: "Gemma 4 26B", |
|
591 |
+ contextWindow: 262_144, maxOutputTokens: 32_768, |
|
592 |
+ capabilities: ModelCapabilities(jsonMode: true), |
|
593 |
+ pricing: nil, |
|
594 |
+ parameterSupport: ParameterSupport() |
|
595 |
+ ), |
|
596 |
+ AIModel( |
|
597 |
+ id: "gemma-4-31b-it", provider: .gemini, displayName: "Gemma 4 31B", |
|
598 |
+ contextWindow: 262_144, maxOutputTokens: 32_768, |
|
599 |
+ capabilities: ModelCapabilities(jsonMode: true), |
|
600 |
+ pricing: nil, |
|
601 |
+ parameterSupport: ParameterSupport() |
|
602 |
+ ), |
| 40 |
603 |
] |
|
604 |
+ |
|
605 |
+ // MARK: - Alibaba Qwen (DashScope) |
|
606 |
+ |
|
607 |
+ static let qwen: [AIModel] = [ |
|
608 |
+ // Flagship commercial |
|
609 |
+ AIModel( |
|
610 |
+ id: "qwen3.7-max", provider: .qwen, displayName: "Qwen3.7 Max", |
|
611 |
+ contextWindow: 1_000_000, maxOutputTokens: nil, |
|
612 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
613 |
+ pricing: ModelPricing(inputPerMTok: 2.50, outputPerMTok: 7.50), |
|
614 |
+ parameterSupport: ParameterSupport(thinkingToggle: true), |
|
615 |
+ isRecommended: true |
|
616 |
+ ), |
|
617 |
+ AIModel( |
|
618 |
+ id: "qwen3.7-plus", provider: .qwen, displayName: "Qwen3.7 Plus", |
|
619 |
+ contextWindow: 1_000_000, maxOutputTokens: 65_536, |
|
620 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
621 |
+ pricing: ModelPricing(inputPerMTok: 0.32, outputPerMTok: 1.28), |
|
622 |
+ parameterSupport: ParameterSupport(thinkingToggle: true), |
|
623 |
+ isRecommended: true |
|
624 |
+ ), |
|
625 |
+ AIModel( |
|
626 |
+ id: "qwen3.7-flash", provider: .qwen, displayName: "Qwen3.7 Flash", |
|
627 |
+ contextWindow: 1_000_000, maxOutputTokens: 65_536, |
|
628 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
629 |
+ pricing: ModelPricing(inputPerMTok: 0.03, outputPerMTok: 0.13), |
|
630 |
+ parameterSupport: ParameterSupport(thinkingToggle: true), |
|
631 |
+ isRecommended: true |
|
632 |
+ ), |
|
633 |
+ AIModel( |
|
634 |
+ id: "qwen3.6-plus", provider: .qwen, displayName: "Qwen3.6 Plus", |
|
635 |
+ contextWindow: 1_000_000, maxOutputTokens: nil, |
|
636 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
637 |
+ pricing: nil, |
|
638 |
+ parameterSupport: ParameterSupport(thinkingToggle: true) |
|
639 |
+ ), |
|
640 |
+ AIModel( |
|
641 |
+ id: "qwen3.6-flash", provider: .qwen, displayName: "Qwen3.6 Flash", |
|
642 |
+ contextWindow: 1_000_000, maxOutputTokens: nil, |
|
643 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
644 |
+ pricing: nil, |
|
645 |
+ parameterSupport: ParameterSupport(thinkingToggle: true) |
|
646 |
+ ), |
|
647 |
+ AIModel( |
|
648 |
+ id: "qwen3.5-plus", provider: .qwen, displayName: "Qwen3.5 Plus", |
|
649 |
+ contextWindow: 1_000_000, maxOutputTokens: nil, |
|
650 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
651 |
+ pricing: nil, |
|
652 |
+ parameterSupport: ParameterSupport(thinkingToggle: true) |
|
653 |
+ ), |
|
654 |
+ AIModel( |
|
655 |
+ id: "qwen3.5-flash", provider: .qwen, displayName: "Qwen3.5 Flash", |
|
656 |
+ contextWindow: 1_000_000, maxOutputTokens: nil, |
|
657 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
658 |
+ pricing: nil, |
|
659 |
+ parameterSupport: ParameterSupport(thinkingToggle: true) |
|
660 |
+ ), |
|
661 |
+ // Stable aliases (previous-gen commercial) |
|
662 |
+ AIModel( |
|
663 |
+ id: "qwen-max", provider: .qwen, displayName: "Qwen Max", |
|
664 |
+ contextWindow: 128_000, maxOutputTokens: nil, |
|
665 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
666 |
+ pricing: nil, |
|
667 |
+ parameterSupport: ParameterSupport(thinkingToggle: true) |
|
668 |
+ ), |
|
669 |
+ AIModel( |
|
670 |
+ id: "qwen-plus", provider: .qwen, displayName: "Qwen Plus", |
|
671 |
+ contextWindow: 1_000_000, maxOutputTokens: nil, |
|
672 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
673 |
+ pricing: nil, |
|
674 |
+ parameterSupport: ParameterSupport(thinkingToggle: true) |
|
675 |
+ ), |
|
676 |
+ AIModel( |
|
677 |
+ id: "qwen-turbo", provider: .qwen, displayName: "Qwen Turbo", |
|
678 |
+ contextWindow: 1_000_000, maxOutputTokens: nil, |
|
679 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
680 |
+ pricing: nil, |
|
681 |
+ parameterSupport: ParameterSupport(thinkingToggle: true), |
|
682 |
+ isLegacy: true |
|
683 |
+ ), |
|
684 |
+ AIModel( |
|
685 |
+ id: "qwen-flash", provider: .qwen, displayName: "Qwen Flash", |
|
686 |
+ contextWindow: 1_000_000, maxOutputTokens: nil, |
|
687 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
688 |
+ pricing: nil, |
|
689 |
+ parameterSupport: ParameterSupport(thinkingToggle: true) |
|
690 |
+ ), |
|
691 |
+ // Coder family |
|
692 |
+ AIModel( |
|
693 |
+ id: "qwen3-coder-plus", provider: .qwen, displayName: "Qwen3 Coder Plus", |
|
694 |
+ contextWindow: 1_000_000, maxOutputTokens: nil, |
|
695 |
+ capabilities: ModelCapabilities(tools: true, jsonMode: true), |
|
696 |
+ pricing: nil, |
|
697 |
+ parameterSupport: ParameterSupport() |
|
698 |
+ ), |
|
699 |
+ AIModel( |
|
700 |
+ id: "qwen3-coder-flash", provider: .qwen, displayName: "Qwen3 Coder Flash", |
|
701 |
+ contextWindow: 1_000_000, maxOutputTokens: nil, |
|
702 |
+ capabilities: ModelCapabilities(tools: true, jsonMode: true), |
|
703 |
+ pricing: nil, |
|
704 |
+ parameterSupport: ParameterSupport() |
|
705 |
+ ), |
|
706 |
+ AIModel( |
|
707 |
+ id: "qwen3-coder-next", provider: .qwen, displayName: "Qwen3 Coder Next", |
|
708 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
709 |
+ capabilities: ModelCapabilities(tools: true, jsonMode: true), |
|
710 |
+ pricing: nil, |
|
711 |
+ parameterSupport: ParameterSupport() |
|
712 |
+ ), |
|
713 |
+ AIModel( |
|
714 |
+ id: "qwen3-coder-480b-a35b-instruct", provider: .qwen, displayName: "Qwen3 Coder 480B A35B", |
|
715 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
716 |
+ capabilities: ModelCapabilities(tools: true, jsonMode: true), |
|
717 |
+ pricing: nil, |
|
718 |
+ parameterSupport: ParameterSupport() |
|
719 |
+ ), |
|
720 |
+ // Vision-language |
|
721 |
+ AIModel( |
|
722 |
+ id: "qwen3-vl-plus", provider: .qwen, displayName: "Qwen3 VL Plus", |
|
723 |
+ contextWindow: 1_000_000, maxOutputTokens: 65_536, |
|
724 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
725 |
+ pricing: nil, |
|
726 |
+ parameterSupport: ParameterSupport(thinkingToggle: true) |
|
727 |
+ ), |
|
728 |
+ AIModel( |
|
729 |
+ id: "qwen3-vl-flash", provider: .qwen, displayName: "Qwen3 VL Flash", |
|
730 |
+ contextWindow: 1_000_000, maxOutputTokens: 65_536, |
|
731 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
732 |
+ pricing: nil, |
|
733 |
+ parameterSupport: ParameterSupport(thinkingToggle: true) |
|
734 |
+ ), |
|
735 |
+ AIModel( |
|
736 |
+ id: "qwen3-vl-235b-a22b-instruct", provider: .qwen, displayName: "Qwen3 VL 235B Instruct", |
|
737 |
+ contextWindow: 131_072, maxOutputTokens: nil, |
|
738 |
+ capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
|
739 |
+ pricing: nil, |
|
740 |
+ parameterSupport: ParameterSupport() |
|
741 |
+ ), |
|
742 |
+ AIModel( |
|
743 |
+ id: "qwen3-vl-235b-a22b-thinking", provider: .qwen, displayName: "Qwen3 VL 235B Thinking", |
|
744 |
+ contextWindow: 131_072, maxOutputTokens: nil, |
|
745 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
746 |
+ pricing: nil, |
|
747 |
+ parameterSupport: ParameterSupport() |
|
748 |
+ ), |
|
749 |
+ AIModel( |
|
750 |
+ id: "qvq-max", provider: .qwen, displayName: "QVQ Max", |
|
751 |
+ contextWindow: 131_072, maxOutputTokens: nil, |
|
752 |
+ capabilities: ModelCapabilities(vision: true, reasoning: true, jsonMode: true), |
|
753 |
+ pricing: nil, |
|
754 |
+ parameterSupport: ParameterSupport() |
|
755 |
+ ), |
|
756 |
+ // Reasoning-only |
|
757 |
+ AIModel( |
|
758 |
+ id: "qwq-plus", provider: .qwen, displayName: "QwQ Plus", |
|
759 |
+ contextWindow: 131_072, maxOutputTokens: nil, |
|
760 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
761 |
+ pricing: nil, |
|
762 |
+ parameterSupport: ParameterSupport() |
|
763 |
+ ), |
|
764 |
+ // Open-weights Qwen hosted on DashScope |
|
765 |
+ AIModel( |
|
766 |
+ id: "qwen3.5-397b-a17b", provider: .qwen, displayName: "Qwen3.5 397B A17B", |
|
767 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
768 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
769 |
+ pricing: nil, |
|
770 |
+ parameterSupport: ParameterSupport(thinkingToggle: true) |
|
771 |
+ ), |
|
772 |
+ AIModel( |
|
773 |
+ id: "qwen3.5-122b-a10b", provider: .qwen, displayName: "Qwen3.5 122B A10B", |
|
774 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
775 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
776 |
+ pricing: nil, |
|
777 |
+ parameterSupport: ParameterSupport(thinkingToggle: true) |
|
778 |
+ ), |
|
779 |
+ AIModel( |
|
780 |
+ id: "qwen3.5-35b-a3b", provider: .qwen, displayName: "Qwen3.5 35B A3B", |
|
781 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
782 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
783 |
+ pricing: nil, |
|
784 |
+ parameterSupport: ParameterSupport(thinkingToggle: true) |
|
785 |
+ ), |
|
786 |
+ AIModel( |
|
787 |
+ id: "qwen3-235b-a22b-instruct-2507", provider: .qwen, displayName: "Qwen3 235B Instruct 2507", |
|
788 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
789 |
+ capabilities: ModelCapabilities(tools: true, jsonMode: true), |
|
790 |
+ pricing: nil, |
|
791 |
+ parameterSupport: ParameterSupport() |
|
792 |
+ ), |
|
793 |
+ AIModel( |
|
794 |
+ id: "qwen3-235b-a22b-thinking-2507", provider: .qwen, displayName: "Qwen3 235B Thinking 2507", |
|
795 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
796 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
797 |
+ pricing: nil, |
|
798 |
+ parameterSupport: ParameterSupport() |
|
799 |
+ ), |
|
800 |
+ AIModel( |
|
801 |
+ id: "qwen3-next-80b-a3b-instruct", provider: .qwen, displayName: "Qwen3 Next 80B Instruct", |
|
802 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
803 |
+ capabilities: ModelCapabilities(tools: true, jsonMode: true), |
|
804 |
+ pricing: nil, |
|
805 |
+ parameterSupport: ParameterSupport() |
|
806 |
+ ), |
|
807 |
+ AIModel( |
|
808 |
+ id: "qwen3-next-80b-a3b-thinking", provider: .qwen, displayName: "Qwen3 Next 80B Thinking", |
|
809 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
810 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
811 |
+ pricing: nil, |
|
812 |
+ parameterSupport: ParameterSupport() |
|
813 |
+ ), |
|
814 |
+ // Third-party models hosted on DashScope |
|
815 |
+ AIModel( |
|
816 |
+ id: "deepseek-v4-pro", provider: .qwen, displayName: "DeepSeek V4 Pro (DashScope)", |
|
817 |
+ contextWindow: 1_000_000, maxOutputTokens: nil, |
|
818 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
819 |
+ pricing: nil, |
|
820 |
+ parameterSupport: ParameterSupport(thinkingToggle: true) |
|
821 |
+ ), |
|
822 |
+ AIModel( |
|
823 |
+ id: "deepseek-v4-flash", provider: .qwen, displayName: "DeepSeek V4 Flash (DashScope)", |
|
824 |
+ contextWindow: 1_000_000, maxOutputTokens: nil, |
|
825 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
826 |
+ pricing: nil, |
|
827 |
+ parameterSupport: ParameterSupport(thinkingToggle: true) |
|
828 |
+ ), |
|
829 |
+ AIModel( |
|
830 |
+ id: "glm-5.2", provider: .qwen, displayName: "GLM 5.2 (DashScope)", |
|
831 |
+ contextWindow: 198_000, maxOutputTokens: nil, |
|
832 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
833 |
+ pricing: nil, |
|
834 |
+ parameterSupport: ParameterSupport(thinkingToggle: true) |
|
835 |
+ ), |
|
836 |
+ AIModel( |
|
837 |
+ id: "kimi-k2.7-code", provider: .qwen, displayName: "Kimi K2.7 Code (DashScope)", |
|
838 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
839 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
840 |
+ pricing: nil, |
|
841 |
+ parameterSupport: ParameterSupport(thinkingToggle: true) |
|
842 |
+ ), |
|
843 |
+ ] |
|
844 |
+ |
|
845 |
+ // MARK: - DeepSeek |
|
846 |
+ |
|
847 |
+ static let deepseek: [AIModel] = [ |
|
848 |
+ AIModel( |
|
849 |
+ id: "deepseek-v4-flash", provider: .deepseek, displayName: "DeepSeek V4 Flash", |
|
850 |
+ contextWindow: 1_000_000, maxOutputTokens: 384_000, |
|
851 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
852 |
+ pricing: ModelPricing(inputPerMTok: 0.14, outputPerMTok: 0.28), |
|
853 |
+ parameterSupport: ParameterSupport(reasoningEffort: true, thinkingToggle: true), |
|
854 |
+ isRecommended: true |
|
855 |
+ ), |
|
856 |
+ AIModel( |
|
857 |
+ id: "deepseek-v4-pro", provider: .deepseek, displayName: "DeepSeek V4 Pro", |
|
858 |
+ contextWindow: 1_000_000, maxOutputTokens: 384_000, |
|
859 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
860 |
+ pricing: ModelPricing(inputPerMTok: 0.435, outputPerMTok: 0.87), |
|
861 |
+ parameterSupport: ParameterSupport(reasoningEffort: true, thinkingToggle: true), |
|
862 |
+ isRecommended: true |
|
863 |
+ ), |
|
864 |
+ ] |
|
865 |
+ |
|
866 |
+ // MARK: - Kimi (Moonshot AI) |
|
867 |
+ |
|
868 |
+ static let kimi: [AIModel] = [ |
|
869 |
+ AIModel( |
|
870 |
+ id: "kimi-k3", provider: .kimi, displayName: "Kimi K3", |
|
871 |
+ contextWindow: 1_048_576, maxOutputTokens: 131_072, |
|
872 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
873 |
+ pricing: ModelPricing(inputPerMTok: 3.00, outputPerMTok: 15.00), |
|
874 |
+ // Thinking always on with preserved thinking; depth via reasoning_effort. |
|
875 |
+ parameterSupport: ParameterSupport(temperature: false, topP: false, usesMaxCompletionTokens: true, reasoningEffort: true), |
|
876 |
+ isRecommended: true |
|
877 |
+ ), |
|
878 |
+ AIModel( |
|
879 |
+ id: "kimi-k2.7-code", provider: .kimi, displayName: "Kimi K2.7 Code", |
|
880 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
881 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
882 |
+ pricing: ModelPricing(inputPerMTok: 0.95, outputPerMTok: 4.00), |
|
883 |
+ parameterSupport: ParameterSupport(temperature: false, topP: false, usesMaxCompletionTokens: true), |
|
884 |
+ isRecommended: true |
|
885 |
+ ), |
|
886 |
+ AIModel( |
|
887 |
+ id: "kimi-k2.7-code-highspeed", provider: .kimi, displayName: "Kimi K2.7 Code Highspeed", |
|
888 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
889 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
890 |
+ pricing: ModelPricing(inputPerMTok: 1.90, outputPerMTok: 8.00), |
|
891 |
+ parameterSupport: ParameterSupport(temperature: false, topP: false, usesMaxCompletionTokens: true) |
|
892 |
+ ), |
|
893 |
+ AIModel( |
|
894 |
+ id: "kimi-k2.6", provider: .kimi, displayName: "Kimi K2.6", |
|
895 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
896 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
897 |
+ pricing: ModelPricing(inputPerMTok: 0.95, outputPerMTok: 4.00), |
|
898 |
+ parameterSupport: ParameterSupport(temperature: false, topP: false, usesMaxCompletionTokens: true, thinkingToggle: true) |
|
899 |
+ ), |
|
900 |
+ AIModel( |
|
901 |
+ id: "kimi-k2.5", provider: .kimi, displayName: "Kimi K2.5", |
|
902 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
903 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
904 |
+ pricing: ModelPricing(inputPerMTok: 0.60, outputPerMTok: 3.00), |
|
905 |
+ parameterSupport: ParameterSupport(temperature: false, topP: false, usesMaxCompletionTokens: true, thinkingToggle: true) |
|
906 |
+ ), |
|
907 |
+ // Legacy moonshot-v1 "classic" series (temperature capped at 1.0) |
|
908 |
+ AIModel( |
|
909 |
+ id: "moonshot-v1-8k", provider: .kimi, displayName: "Moonshot v1 8K", |
|
910 |
+ contextWindow: 8_192, maxOutputTokens: nil, |
|
911 |
+ capabilities: ModelCapabilities(tools: true, jsonMode: true), |
|
912 |
+ pricing: ModelPricing(inputPerMTok: 0.20, outputPerMTok: 2.00), |
|
913 |
+ parameterSupport: .openAIDefault, |
|
914 |
+ isLegacy: true |
|
915 |
+ ), |
|
916 |
+ AIModel( |
|
917 |
+ id: "moonshot-v1-32k", provider: .kimi, displayName: "Moonshot v1 32K", |
|
918 |
+ contextWindow: 32_768, maxOutputTokens: nil, |
|
919 |
+ capabilities: ModelCapabilities(tools: true, jsonMode: true), |
|
920 |
+ pricing: ModelPricing(inputPerMTok: 1.00, outputPerMTok: 3.00), |
|
921 |
+ parameterSupport: .openAIDefault, |
|
922 |
+ isLegacy: true |
|
923 |
+ ), |
|
924 |
+ AIModel( |
|
925 |
+ id: "moonshot-v1-128k", provider: .kimi, displayName: "Moonshot v1 128K", |
|
926 |
+ contextWindow: 131_072, maxOutputTokens: nil, |
|
927 |
+ capabilities: ModelCapabilities(tools: true, jsonMode: true), |
|
928 |
+ pricing: ModelPricing(inputPerMTok: 2.00, outputPerMTok: 5.00), |
|
929 |
+ parameterSupport: .openAIDefault, |
|
930 |
+ isLegacy: true |
|
931 |
+ ), |
|
932 |
+ AIModel( |
|
933 |
+ id: "moonshot-v1-auto", provider: .kimi, displayName: "Moonshot v1 Auto", |
|
934 |
+ contextWindow: 131_072, maxOutputTokens: nil, |
|
935 |
+ capabilities: ModelCapabilities(tools: true, jsonMode: true), |
|
936 |
+ pricing: nil, |
|
937 |
+ parameterSupport: .openAIDefault, |
|
938 |
+ isLegacy: true |
|
939 |
+ ), |
|
940 |
+ AIModel( |
|
941 |
+ id: "moonshot-v1-8k-vision-preview", provider: .kimi, displayName: "Moonshot v1 8K Vision", |
|
942 |
+ contextWindow: 8_192, maxOutputTokens: nil, |
|
943 |
+ capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
|
944 |
+ pricing: ModelPricing(inputPerMTok: 0.20, outputPerMTok: 2.00), |
|
945 |
+ parameterSupport: .openAIDefault, |
|
946 |
+ isLegacy: true |
|
947 |
+ ), |
|
948 |
+ AIModel( |
|
949 |
+ id: "moonshot-v1-32k-vision-preview", provider: .kimi, displayName: "Moonshot v1 32K Vision", |
|
950 |
+ contextWindow: 32_768, maxOutputTokens: nil, |
|
951 |
+ capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
|
952 |
+ pricing: ModelPricing(inputPerMTok: 1.00, outputPerMTok: 3.00), |
|
953 |
+ parameterSupport: .openAIDefault, |
|
954 |
+ isLegacy: true |
|
955 |
+ ), |
|
956 |
+ AIModel( |
|
957 |
+ id: "moonshot-v1-128k-vision-preview", provider: .kimi, displayName: "Moonshot v1 128K Vision", |
|
958 |
+ contextWindow: 131_072, maxOutputTokens: nil, |
|
959 |
+ capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
|
960 |
+ pricing: ModelPricing(inputPerMTok: 2.00, outputPerMTok: 5.00), |
|
961 |
+ parameterSupport: .openAIDefault, |
|
962 |
+ isLegacy: true |
|
963 |
+ ), |
|
964 |
+ ] |
|
965 |
+ |
|
966 |
+ // MARK: - Perplexity |
|
967 |
+ |
|
968 |
+ static let perplexity: [AIModel] = [ |
|
969 |
+ AIModel( |
|
970 |
+ id: "sonar", provider: .perplexity, displayName: "Sonar", |
|
971 |
+ contextWindow: 128_000, maxOutputTokens: 128_000, |
|
972 |
+ capabilities: ModelCapabilities(jsonMode: true, citations: true), |
|
973 |
+ pricing: ModelPricing(inputPerMTok: 1.00, outputPerMTok: 1.00), |
|
974 |
+ parameterSupport: ParameterSupport(), |
|
975 |
+ isRecommended: true |
|
976 |
+ ), |
|
977 |
+ AIModel( |
|
978 |
+ id: "sonar-pro", provider: .perplexity, displayName: "Sonar Pro", |
|
979 |
+ contextWindow: 200_000, maxOutputTokens: 8_000, |
|
980 |
+ capabilities: ModelCapabilities(jsonMode: true, citations: true), |
|
981 |
+ pricing: ModelPricing(inputPerMTok: 3.00, outputPerMTok: 15.00), |
|
982 |
+ parameterSupport: ParameterSupport(), |
|
983 |
+ isRecommended: true |
|
984 |
+ ), |
|
985 |
+ AIModel( |
|
986 |
+ id: "sonar-reasoning-pro", provider: .perplexity, displayName: "Sonar Reasoning Pro", |
|
987 |
+ contextWindow: 128_000, maxOutputTokens: nil, |
|
988 |
+ capabilities: ModelCapabilities(reasoning: true, jsonMode: true, citations: true), |
|
989 |
+ pricing: ModelPricing(inputPerMTok: 2.00, outputPerMTok: 8.00), |
|
990 |
+ parameterSupport: ParameterSupport(reasoningEffort: true) |
|
991 |
+ ), |
|
992 |
+ AIModel( |
|
993 |
+ id: "sonar-deep-research", provider: .perplexity, displayName: "Sonar Deep Research", |
|
994 |
+ contextWindow: 128_000, maxOutputTokens: nil, |
|
995 |
+ capabilities: ModelCapabilities(reasoning: true, jsonMode: true, citations: true), |
|
996 |
+ pricing: ModelPricing(inputPerMTok: 2.00, outputPerMTok: 8.00), |
|
997 |
+ parameterSupport: ParameterSupport(reasoningEffort: true) |
|
998 |
+ ), |
|
999 |
+ ] |
|
1000 |
+ |
|
1001 |
+ // MARK: - Together AI |
|
1002 |
+ |
|
1003 |
+ static let together: [AIModel] = [ |
|
1004 |
+ AIModel( |
|
1005 |
+ id: "moonshotai/Kimi-K3", provider: .together, displayName: "Kimi K3", |
|
1006 |
+ contextWindow: 1_000_000, maxOutputTokens: nil, |
|
1007 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1008 |
+ pricing: ModelPricing(inputPerMTok: 3.00, outputPerMTok: 15.00), |
|
1009 |
+ parameterSupport: .openAIDefault, |
|
1010 |
+ isRecommended: true |
|
1011 |
+ ), |
|
1012 |
+ AIModel( |
|
1013 |
+ id: "moonshotai/Kimi-K2.7-Code", provider: .together, displayName: "Kimi K2.7 Code", |
|
1014 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
1015 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1016 |
+ pricing: ModelPricing(inputPerMTok: 0.95, outputPerMTok: 4.00), |
|
1017 |
+ parameterSupport: .openAIDefault |
|
1018 |
+ ), |
|
1019 |
+ AIModel( |
|
1020 |
+ id: "moonshotai/Kimi-K2.6", provider: .together, displayName: "Kimi K2.6", |
|
1021 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
1022 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1023 |
+ pricing: ModelPricing(inputPerMTok: 1.20, outputPerMTok: 4.50), |
|
1024 |
+ parameterSupport: .openAIDefault |
|
1025 |
+ ), |
|
1026 |
+ AIModel( |
|
1027 |
+ id: "deepseek-ai/DeepSeek-V4-Pro", provider: .together, displayName: "DeepSeek V4 Pro", |
|
1028 |
+ contextWindow: 512_000, maxOutputTokens: nil, |
|
1029 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1030 |
+ pricing: ModelPricing(inputPerMTok: 1.74, outputPerMTok: 3.48), |
|
1031 |
+ parameterSupport: .openAIDefault, |
|
1032 |
+ isRecommended: true |
|
1033 |
+ ), |
|
1034 |
+ AIModel( |
|
1035 |
+ id: "deepseek-ai/DeepSeek-V3.1", provider: .together, displayName: "DeepSeek V3.1", |
|
1036 |
+ contextWindow: 131_072, maxOutputTokens: nil, |
|
1037 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1038 |
+ pricing: ModelPricing(inputPerMTok: 0.60, outputPerMTok: 1.70), |
|
1039 |
+ parameterSupport: .openAIDefault |
|
1040 |
+ ), |
|
1041 |
+ AIModel( |
|
1042 |
+ id: "deepseek-ai/DeepSeek-R1-0528", provider: .together, displayName: "DeepSeek R1 0528", |
|
1043 |
+ contextWindow: 163_840, maxOutputTokens: nil, |
|
1044 |
+ capabilities: ModelCapabilities(reasoning: true), |
|
1045 |
+ pricing: ModelPricing(inputPerMTok: 3.00, outputPerMTok: 7.00), |
|
1046 |
+ parameterSupport: .openAIDefault |
|
1047 |
+ ), |
|
1048 |
+ AIModel( |
|
1049 |
+ id: "deepseek-ai/DeepSeek-R1-Distill-Llama-70B", provider: .together, displayName: "DeepSeek R1 Distill Llama 70B", |
|
1050 |
+ contextWindow: 131_072, maxOutputTokens: nil, |
|
1051 |
+ capabilities: ModelCapabilities(reasoning: true), |
|
1052 |
+ pricing: ModelPricing(inputPerMTok: 2.00, outputPerMTok: 2.00), |
|
1053 |
+ parameterSupport: .openAIDefault |
|
1054 |
+ ), |
|
1055 |
+ AIModel( |
|
1056 |
+ id: "zai-org/GLM-5.2", provider: .together, displayName: "GLM 5.2", |
|
1057 |
+ contextWindow: 512_000, maxOutputTokens: nil, |
|
1058 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1059 |
+ pricing: ModelPricing(inputPerMTok: 1.40, outputPerMTok: 4.40), |
|
1060 |
+ parameterSupport: .openAIDefault |
|
1061 |
+ ), |
|
1062 |
+ AIModel( |
|
1063 |
+ id: "zai-org/GLM-5.1", provider: .together, displayName: "GLM 5.1", |
|
1064 |
+ contextWindow: 202_752, maxOutputTokens: nil, |
|
1065 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1066 |
+ pricing: ModelPricing(inputPerMTok: 1.40, outputPerMTok: 4.40), |
|
1067 |
+ parameterSupport: .openAIDefault |
|
1068 |
+ ), |
|
1069 |
+ AIModel( |
|
1070 |
+ id: "zai-org/GLM-5", provider: .together, displayName: "GLM 5", |
|
1071 |
+ contextWindow: 202_752, maxOutputTokens: nil, |
|
1072 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1073 |
+ pricing: ModelPricing(inputPerMTok: 1.00, outputPerMTok: 3.20), |
|
1074 |
+ parameterSupport: .openAIDefault |
|
1075 |
+ ), |
|
1076 |
+ AIModel( |
|
1077 |
+ id: "zai-org/GLM-4.7", provider: .together, displayName: "GLM 4.7", |
|
1078 |
+ contextWindow: 202_752, maxOutputTokens: nil, |
|
1079 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1080 |
+ pricing: ModelPricing(inputPerMTok: 0.45, outputPerMTok: 2.00), |
|
1081 |
+ parameterSupport: .openAIDefault |
|
1082 |
+ ), |
|
1083 |
+ AIModel( |
|
1084 |
+ id: "Qwen/Qwen3.7-Max", provider: .together, displayName: "Qwen3.7 Max", |
|
1085 |
+ contextWindow: 1_000_000, maxOutputTokens: nil, |
|
1086 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1087 |
+ pricing: ModelPricing(inputPerMTok: 1.25, outputPerMTok: 3.75), |
|
1088 |
+ parameterSupport: .openAIDefault |
|
1089 |
+ ), |
|
1090 |
+ AIModel( |
|
1091 |
+ id: "Qwen/Qwen3.7-Plus", provider: .together, displayName: "Qwen3.7 Plus", |
|
1092 |
+ contextWindow: 1_000_000, maxOutputTokens: nil, |
|
1093 |
+ capabilities: ModelCapabilities(tools: true, jsonMode: true), |
|
1094 |
+ pricing: ModelPricing(inputPerMTok: 0.32, outputPerMTok: 1.28), |
|
1095 |
+ parameterSupport: .openAIDefault |
|
1096 |
+ ), |
|
1097 |
+ AIModel( |
|
1098 |
+ id: "Qwen/Qwen3.6-Plus", provider: .together, displayName: "Qwen3.6 Plus", |
|
1099 |
+ contextWindow: 1_000_000, maxOutputTokens: nil, |
|
1100 |
+ capabilities: ModelCapabilities(tools: true, jsonMode: true), |
|
1101 |
+ pricing: ModelPricing(inputPerMTok: 0.50, outputPerMTok: 3.00), |
|
1102 |
+ parameterSupport: .openAIDefault |
|
1103 |
+ ), |
|
1104 |
+ AIModel( |
|
1105 |
+ id: "Qwen/Qwen3.5-397B-A17B", provider: .together, displayName: "Qwen3.5 397B A17B", |
|
1106 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
1107 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1108 |
+ pricing: ModelPricing(inputPerMTok: 0.60, outputPerMTok: 3.60), |
|
1109 |
+ parameterSupport: .openAIDefault |
|
1110 |
+ ), |
|
1111 |
+ AIModel( |
|
1112 |
+ id: "Qwen/Qwen3.5-9B", provider: .together, displayName: "Qwen3.5 9B", |
|
1113 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
1114 |
+ capabilities: ModelCapabilities(tools: true, jsonMode: true), |
|
1115 |
+ pricing: ModelPricing(inputPerMTok: 0.17, outputPerMTok: 0.25), |
|
1116 |
+ parameterSupport: .openAIDefault |
|
1117 |
+ ), |
|
1118 |
+ AIModel( |
|
1119 |
+ id: "Qwen/Qwen3-Next-80B-A3B-Instruct", provider: .together, displayName: "Qwen3 Next 80B Instruct", |
|
1120 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
1121 |
+ capabilities: ModelCapabilities(tools: true, jsonMode: true), |
|
1122 |
+ pricing: ModelPricing(inputPerMTok: 0.15, outputPerMTok: 1.50), |
|
1123 |
+ parameterSupport: .openAIDefault |
|
1124 |
+ ), |
|
1125 |
+ AIModel( |
|
1126 |
+ id: "Qwen/Qwen3-Next-80B-A3B-Thinking", provider: .together, displayName: "Qwen3 Next 80B Thinking", |
|
1127 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
1128 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1129 |
+ pricing: ModelPricing(inputPerMTok: 0.15, outputPerMTok: 1.50), |
|
1130 |
+ parameterSupport: .openAIDefault |
|
1131 |
+ ), |
|
1132 |
+ AIModel( |
|
1133 |
+ id: "Qwen/Qwen3-Coder-Next-FP8", provider: .together, displayName: "Qwen3 Coder Next", |
|
1134 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
1135 |
+ capabilities: ModelCapabilities(tools: true, jsonMode: true), |
|
1136 |
+ pricing: ModelPricing(inputPerMTok: 0.50, outputPerMTok: 1.20), |
|
1137 |
+ parameterSupport: .openAIDefault |
|
1138 |
+ ), |
|
1139 |
+ AIModel( |
|
1140 |
+ id: "Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8", provider: .together, displayName: "Qwen3 Coder 480B", |
|
1141 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
1142 |
+ capabilities: ModelCapabilities(tools: true, jsonMode: true), |
|
1143 |
+ pricing: ModelPricing(inputPerMTok: 2.00, outputPerMTok: 2.00), |
|
1144 |
+ parameterSupport: .openAIDefault |
|
1145 |
+ ), |
|
1146 |
+ AIModel( |
|
1147 |
+ id: "Qwen/Qwen3-VL-32B-Instruct", provider: .together, displayName: "Qwen3 VL 32B", |
|
1148 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
1149 |
+ capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
|
1150 |
+ pricing: ModelPricing(inputPerMTok: 0.50, outputPerMTok: 1.50), |
|
1151 |
+ parameterSupport: .openAIDefault |
|
1152 |
+ ), |
|
1153 |
+ AIModel( |
|
1154 |
+ id: "Qwen/Qwen3-VL-8B-Instruct", provider: .together, displayName: "Qwen3 VL 8B", |
|
1155 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
1156 |
+ capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
|
1157 |
+ pricing: ModelPricing(inputPerMTok: 0.18, outputPerMTok: 0.68), |
|
1158 |
+ parameterSupport: .openAIDefault |
|
1159 |
+ ), |
|
1160 |
+ AIModel( |
|
1161 |
+ id: "Qwen/QwQ-32B", provider: .together, displayName: "QwQ 32B", |
|
1162 |
+ contextWindow: 131_072, maxOutputTokens: nil, |
|
1163 |
+ capabilities: ModelCapabilities(reasoning: true), |
|
1164 |
+ pricing: ModelPricing(inputPerMTok: 1.20, outputPerMTok: 1.20), |
|
1165 |
+ parameterSupport: .openAIDefault |
|
1166 |
+ ), |
|
1167 |
+ AIModel( |
|
1168 |
+ id: "meta-llama/Llama-4-Scout-17B-16E-Instruct", provider: .together, displayName: "Llama 4 Scout", |
|
1169 |
+ contextWindow: 1_048_576, maxOutputTokens: nil, |
|
1170 |
+ capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
|
1171 |
+ pricing: ModelPricing(inputPerMTok: 0.18, outputPerMTok: 0.59), |
|
1172 |
+ parameterSupport: .openAIDefault |
|
1173 |
+ ), |
|
1174 |
+ AIModel( |
|
1175 |
+ id: "meta-llama/Llama-3.3-70B-Instruct-Turbo", provider: .together, displayName: "Llama 3.3 70B Turbo", |
|
1176 |
+ contextWindow: 131_072, maxOutputTokens: nil, |
|
1177 |
+ capabilities: ModelCapabilities(tools: true, jsonMode: true), |
|
1178 |
+ pricing: ModelPricing(inputPerMTok: 1.04, outputPerMTok: 1.04), |
|
1179 |
+ parameterSupport: .openAIDefault |
|
1180 |
+ ), |
|
1181 |
+ AIModel( |
|
1182 |
+ id: "meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo", provider: .together, displayName: "Llama 3.1 8B Turbo", |
|
1183 |
+ contextWindow: 131_072, maxOutputTokens: nil, |
|
1184 |
+ capabilities: ModelCapabilities(tools: true, jsonMode: true), |
|
1185 |
+ pricing: ModelPricing(inputPerMTok: 0.18, outputPerMTok: 0.18), |
|
1186 |
+ parameterSupport: .openAIDefault |
|
1187 |
+ ), |
|
1188 |
+ AIModel( |
|
1189 |
+ id: "openai/gpt-oss-120b", provider: .together, displayName: "GPT-OSS 120B", |
|
1190 |
+ contextWindow: 131_072, maxOutputTokens: nil, |
|
1191 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1192 |
+ pricing: ModelPricing(inputPerMTok: 0.15, outputPerMTok: 0.60), |
|
1193 |
+ parameterSupport: ParameterSupport(frequencyPenalty: true, presencePenalty: true, reasoningEffort: true), |
|
1194 |
+ isRecommended: true |
|
1195 |
+ ), |
|
1196 |
+ AIModel( |
|
1197 |
+ id: "openai/gpt-oss-20b", provider: .together, displayName: "GPT-OSS 20B", |
|
1198 |
+ contextWindow: 131_072, maxOutputTokens: nil, |
|
1199 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1200 |
+ pricing: ModelPricing(inputPerMTok: 0.05, outputPerMTok: 0.20), |
|
1201 |
+ parameterSupport: ParameterSupport(frequencyPenalty: true, presencePenalty: true, reasoningEffort: true) |
|
1202 |
+ ), |
|
1203 |
+ AIModel( |
|
1204 |
+ id: "nvidia/nemotron-3-ultra-550b-a55b", provider: .together, displayName: "Nemotron 3 Ultra 550B", |
|
1205 |
+ contextWindow: 512_288, maxOutputTokens: nil, |
|
1206 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1207 |
+ pricing: ModelPricing(inputPerMTok: 0.60, outputPerMTok: 3.60), |
|
1208 |
+ parameterSupport: .openAIDefault |
|
1209 |
+ ), |
|
1210 |
+ AIModel( |
|
1211 |
+ id: "MiniMaxAI/MiniMax-M3", provider: .together, displayName: "MiniMax M3", |
|
1212 |
+ contextWindow: 524_288, maxOutputTokens: nil, |
|
1213 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1214 |
+ pricing: ModelPricing(inputPerMTok: 0.30, outputPerMTok: 1.20), |
|
1215 |
+ parameterSupport: .openAIDefault |
|
1216 |
+ ), |
|
1217 |
+ AIModel( |
|
1218 |
+ id: "mistralai/Ministral-3-14B-Instruct-2512", provider: .together, displayName: "Ministral 3 14B", |
|
1219 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
1220 |
+ capabilities: ModelCapabilities(tools: true, jsonMode: true), |
|
1221 |
+ pricing: ModelPricing(inputPerMTok: 0.20, outputPerMTok: 0.20), |
|
1222 |
+ parameterSupport: .openAIDefault |
|
1223 |
+ ), |
|
1224 |
+ AIModel( |
|
1225 |
+ id: "google/gemma-4-31B-it", provider: .together, displayName: "Gemma 4 31B", |
|
1226 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
1227 |
+ capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
|
1228 |
+ pricing: ModelPricing(inputPerMTok: 0.39, outputPerMTok: 0.97), |
|
1229 |
+ parameterSupport: .openAIDefault |
|
1230 |
+ ), |
|
1231 |
+ AIModel( |
|
1232 |
+ id: "thinkingmachines/Inkling", provider: .together, displayName: "Inkling", |
|
1233 |
+ contextWindow: 524_288, maxOutputTokens: nil, |
|
1234 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1235 |
+ pricing: ModelPricing(inputPerMTok: 1.00, outputPerMTok: 4.05), |
|
1236 |
+ parameterSupport: .openAIDefault |
|
1237 |
+ ), |
|
1238 |
+ AIModel( |
|
1239 |
+ id: "mistralai/Mixtral-8x7B-Instruct-v0.1", provider: .together, displayName: "Mixtral 8x7B", |
|
1240 |
+ contextWindow: 32_768, maxOutputTokens: nil, |
|
1241 |
+ capabilities: ModelCapabilities(jsonMode: true), |
|
1242 |
+ pricing: ModelPricing(inputPerMTok: 0.60, outputPerMTok: 0.60), |
|
1243 |
+ parameterSupport: .openAIDefault, |
|
1244 |
+ isLegacy: true |
|
1245 |
+ ), |
|
1246 |
+ ] |
|
1247 |
+ |
|
1248 |
+ // MARK: - DeepInfra |
|
1249 |
+ |
|
1250 |
+ static let deepinfra: [AIModel] = [ |
|
1251 |
+ // Proxied frontier models (Claude / Gemini under DeepInfra billing) |
|
1252 |
+ AIModel( |
|
1253 |
+ id: "anthropic/claude-fable-5", provider: .deepinfra, displayName: "Claude Fable 5", |
|
1254 |
+ contextWindow: 1_000_000, maxOutputTokens: nil, |
|
1255 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
1256 |
+ pricing: ModelPricing(inputPerMTok: 10.00, outputPerMTok: 50.00), |
|
1257 |
+ parameterSupport: ParameterSupport() |
|
1258 |
+ ), |
|
1259 |
+ AIModel( |
|
1260 |
+ id: "anthropic/claude-opus-5", provider: .deepinfra, displayName: "Claude Opus 5", |
|
1261 |
+ contextWindow: 1_000_000, maxOutputTokens: nil, |
|
1262 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
1263 |
+ pricing: ModelPricing(inputPerMTok: 5.00, outputPerMTok: 25.00), |
|
1264 |
+ parameterSupport: ParameterSupport() |
|
1265 |
+ ), |
|
1266 |
+ AIModel( |
|
1267 |
+ id: "anthropic/claude-sonnet-5", provider: .deepinfra, displayName: "Claude Sonnet 5", |
|
1268 |
+ contextWindow: 1_000_000, maxOutputTokens: nil, |
|
1269 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
1270 |
+ pricing: ModelPricing(inputPerMTok: 2.00, outputPerMTok: 10.00), |
|
1271 |
+ parameterSupport: ParameterSupport() |
|
1272 |
+ ), |
|
1273 |
+ AIModel( |
|
1274 |
+ id: "anthropic/claude-opus-4-8", provider: .deepinfra, displayName: "Claude Opus 4.8", |
|
1275 |
+ contextWindow: 1_000_000, maxOutputTokens: nil, |
|
1276 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
1277 |
+ pricing: ModelPricing(inputPerMTok: 5.00, outputPerMTok: 25.00), |
|
1278 |
+ parameterSupport: ParameterSupport() |
|
1279 |
+ ), |
|
1280 |
+ AIModel( |
|
1281 |
+ id: "anthropic/claude-haiku-4-5", provider: .deepinfra, displayName: "Claude Haiku 4.5", |
|
1282 |
+ contextWindow: 200_000, maxOutputTokens: nil, |
|
1283 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
1284 |
+ pricing: ModelPricing(inputPerMTok: 1.00, outputPerMTok: 5.00), |
|
1285 |
+ parameterSupport: ParameterSupport() |
|
1286 |
+ ), |
|
1287 |
+ AIModel( |
|
1288 |
+ id: "google/gemini-3.1-pro", provider: .deepinfra, displayName: "Gemini 3.1 Pro", |
|
1289 |
+ contextWindow: 1_000_000, maxOutputTokens: nil, |
|
1290 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
1291 |
+ pricing: ModelPricing(inputPerMTok: 2.00, outputPerMTok: 12.00), |
|
1292 |
+ parameterSupport: ParameterSupport() |
|
1293 |
+ ), |
|
1294 |
+ AIModel( |
|
1295 |
+ id: "google/gemini-3.5-flash", provider: .deepinfra, displayName: "Gemini 3.5 Flash", |
|
1296 |
+ contextWindow: 1_000_000, maxOutputTokens: nil, |
|
1297 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
1298 |
+ pricing: ModelPricing(inputPerMTok: 1.50, outputPerMTok: 9.00), |
|
1299 |
+ parameterSupport: ParameterSupport() |
|
1300 |
+ ), |
|
1301 |
+ AIModel( |
|
1302 |
+ id: "google/gemini-3.1-flash-lite", provider: .deepinfra, displayName: "Gemini 3.1 Flash-Lite", |
|
1303 |
+ contextWindow: 1_000_000, maxOutputTokens: nil, |
|
1304 |
+ capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
|
1305 |
+ pricing: ModelPricing(inputPerMTok: 0.25, outputPerMTok: 1.50), |
|
1306 |
+ parameterSupport: ParameterSupport() |
|
1307 |
+ ), |
|
1308 |
+ AIModel( |
|
1309 |
+ id: "google/gemini-2.5-pro", provider: .deepinfra, displayName: "Gemini 2.5 Pro", |
|
1310 |
+ contextWindow: 1_000_000, maxOutputTokens: nil, |
|
1311 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
1312 |
+ pricing: ModelPricing(inputPerMTok: 1.25, outputPerMTok: 10.00), |
|
1313 |
+ parameterSupport: ParameterSupport() |
|
1314 |
+ ), |
|
1315 |
+ AIModel( |
|
1316 |
+ id: "google/gemini-2.5-flash", provider: .deepinfra, displayName: "Gemini 2.5 Flash", |
|
1317 |
+ contextWindow: 1_000_000, maxOutputTokens: nil, |
|
1318 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
1319 |
+ pricing: ModelPricing(inputPerMTok: 0.30, outputPerMTok: 2.50), |
|
1320 |
+ parameterSupport: ParameterSupport() |
|
1321 |
+ ), |
|
1322 |
+ // Open-weight chat models |
|
1323 |
+ AIModel( |
|
1324 |
+ id: "deepseek-ai/DeepSeek-V4-Pro", provider: .deepinfra, displayName: "DeepSeek V4 Pro", |
|
1325 |
+ contextWindow: 1_048_576, maxOutputTokens: nil, |
|
1326 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1327 |
+ pricing: ModelPricing(inputPerMTok: 1.30, outputPerMTok: 2.60), |
|
1328 |
+ parameterSupport: .openAIDefault, |
|
1329 |
+ isRecommended: true |
|
1330 |
+ ), |
|
1331 |
+ AIModel( |
|
1332 |
+ id: "deepseek-ai/DeepSeek-V4-Flash", provider: .deepinfra, displayName: "DeepSeek V4 Flash", |
|
1333 |
+ contextWindow: 1_048_576, maxOutputTokens: nil, |
|
1334 |
+ capabilities: ModelCapabilities(tools: true, jsonMode: true), |
|
1335 |
+ pricing: ModelPricing(inputPerMTok: 0.09, outputPerMTok: 0.18), |
|
1336 |
+ parameterSupport: .openAIDefault, |
|
1337 |
+ isRecommended: true |
|
1338 |
+ ), |
|
1339 |
+ AIModel( |
|
1340 |
+ id: "deepseek-ai/DeepSeek-V3.1", provider: .deepinfra, displayName: "DeepSeek V3.1", |
|
1341 |
+ contextWindow: 163_840, maxOutputTokens: nil, |
|
1342 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1343 |
+ pricing: ModelPricing(inputPerMTok: 0.25, outputPerMTok: 0.95), |
|
1344 |
+ parameterSupport: .openAIDefault |
|
1345 |
+ ), |
|
1346 |
+ AIModel( |
|
1347 |
+ id: "deepseek-ai/DeepSeek-R1-0528", provider: .deepinfra, displayName: "DeepSeek R1 0528", |
|
1348 |
+ contextWindow: 163_840, maxOutputTokens: nil, |
|
1349 |
+ capabilities: ModelCapabilities(reasoning: true), |
|
1350 |
+ pricing: ModelPricing(inputPerMTok: 0.50, outputPerMTok: 2.15), |
|
1351 |
+ parameterSupport: .openAIDefault |
|
1352 |
+ ), |
|
1353 |
+ AIModel( |
|
1354 |
+ id: "moonshotai/Kimi-K2.7-Code", provider: .deepinfra, displayName: "Kimi K2.7 Code", |
|
1355 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
1356 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1357 |
+ pricing: ModelPricing(inputPerMTok: 0.74, outputPerMTok: 3.50), |
|
1358 |
+ parameterSupport: .openAIDefault |
|
1359 |
+ ), |
|
1360 |
+ AIModel( |
|
1361 |
+ id: "moonshotai/Kimi-K2.6", provider: .deepinfra, displayName: "Kimi K2.6", |
|
1362 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
1363 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1364 |
+ pricing: ModelPricing(inputPerMTok: 0.75, outputPerMTok: 3.50), |
|
1365 |
+ parameterSupport: .openAIDefault |
|
1366 |
+ ), |
|
1367 |
+ AIModel( |
|
1368 |
+ id: "moonshotai/Kimi-K2.5", provider: .deepinfra, displayName: "Kimi K2.5", |
|
1369 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
1370 |
+ capabilities: ModelCapabilities(tools: true, jsonMode: true), |
|
1371 |
+ pricing: ModelPricing(inputPerMTok: 0.45, outputPerMTok: 2.25), |
|
1372 |
+ parameterSupport: .openAIDefault |
|
1373 |
+ ), |
|
1374 |
+ AIModel( |
|
1375 |
+ id: "zai-org/GLM-5.2", provider: .deepinfra, displayName: "GLM 5.2", |
|
1376 |
+ contextWindow: 1_048_576, maxOutputTokens: nil, |
|
1377 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1378 |
+ pricing: ModelPricing(inputPerMTok: 0.75, outputPerMTok: 2.40), |
|
1379 |
+ parameterSupport: .openAIDefault, |
|
1380 |
+ isRecommended: true |
|
1381 |
+ ), |
|
1382 |
+ AIModel( |
|
1383 |
+ id: "zai-org/GLM-4.7", provider: .deepinfra, displayName: "GLM 4.7", |
|
1384 |
+ contextWindow: 202_752, maxOutputTokens: nil, |
|
1385 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1386 |
+ pricing: ModelPricing(inputPerMTok: 0.40, outputPerMTok: 1.75), |
|
1387 |
+ parameterSupport: .openAIDefault |
|
1388 |
+ ), |
|
1389 |
+ AIModel( |
|
1390 |
+ id: "Qwen/Qwen3.7-Max", provider: .deepinfra, displayName: "Qwen3.7 Max", |
|
1391 |
+ contextWindow: 256_000, maxOutputTokens: nil, |
|
1392 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1393 |
+ pricing: ModelPricing(inputPerMTok: 2.50, outputPerMTok: 7.50), |
|
1394 |
+ parameterSupport: .openAIDefault |
|
1395 |
+ ), |
|
1396 |
+ AIModel( |
|
1397 |
+ id: "Qwen/Qwen3.5-397B-A17B", provider: .deepinfra, displayName: "Qwen3.5 397B A17B", |
|
1398 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
1399 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1400 |
+ pricing: ModelPricing(inputPerMTok: 0.45, outputPerMTok: 3.00), |
|
1401 |
+ parameterSupport: .openAIDefault |
|
1402 |
+ ), |
|
1403 |
+ AIModel( |
|
1404 |
+ id: "Qwen/Qwen3-235B-A22B-Instruct-2507", provider: .deepinfra, displayName: "Qwen3 235B Instruct 2507", |
|
1405 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
1406 |
+ capabilities: ModelCapabilities(tools: true, jsonMode: true), |
|
1407 |
+ pricing: ModelPricing(inputPerMTok: 0.09, outputPerMTok: 0.55), |
|
1408 |
+ parameterSupport: .openAIDefault |
|
1409 |
+ ), |
|
1410 |
+ AIModel( |
|
1411 |
+ id: "Qwen/Qwen3-235B-A22B-Thinking-2507", provider: .deepinfra, displayName: "Qwen3 235B Thinking 2507", |
|
1412 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
1413 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1414 |
+ pricing: ModelPricing(inputPerMTok: 0.23, outputPerMTok: 2.30), |
|
1415 |
+ parameterSupport: .openAIDefault |
|
1416 |
+ ), |
|
1417 |
+ AIModel( |
|
1418 |
+ id: "Qwen/Qwen3-Coder-480B-A35B-Instruct-Turbo", provider: .deepinfra, displayName: "Qwen3 Coder 480B Turbo", |
|
1419 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
1420 |
+ capabilities: ModelCapabilities(tools: true, jsonMode: true), |
|
1421 |
+ pricing: ModelPricing(inputPerMTok: 0.30, outputPerMTok: 1.00), |
|
1422 |
+ parameterSupport: .openAIDefault |
|
1423 |
+ ), |
|
1424 |
+ AIModel( |
|
1425 |
+ id: "Qwen/Qwen3-VL-235B-A22B-Instruct", provider: .deepinfra, displayName: "Qwen3 VL 235B", |
|
1426 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
1427 |
+ capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
|
1428 |
+ pricing: ModelPricing(inputPerMTok: 0.20, outputPerMTok: 0.88), |
|
1429 |
+ parameterSupport: .openAIDefault |
|
1430 |
+ ), |
|
1431 |
+ AIModel( |
|
1432 |
+ id: "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8", provider: .deepinfra, displayName: "Llama 4 Maverick", |
|
1433 |
+ contextWindow: 1_048_576, maxOutputTokens: nil, |
|
1434 |
+ capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
|
1435 |
+ pricing: ModelPricing(inputPerMTok: 0.20, outputPerMTok: 0.80), |
|
1436 |
+ parameterSupport: .openAIDefault |
|
1437 |
+ ), |
|
1438 |
+ AIModel( |
|
1439 |
+ id: "meta-llama/Llama-4-Scout-17B-16E-Instruct", provider: .deepinfra, displayName: "Llama 4 Scout", |
|
1440 |
+ contextWindow: 327_680, maxOutputTokens: nil, |
|
1441 |
+ capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
|
1442 |
+ pricing: ModelPricing(inputPerMTok: 0.10, outputPerMTok: 0.30), |
|
1443 |
+ parameterSupport: .openAIDefault |
|
1444 |
+ ), |
|
1445 |
+ AIModel( |
|
1446 |
+ id: "meta-llama/Llama-3.3-70B-Instruct-Turbo", provider: .deepinfra, displayName: "Llama 3.3 70B Turbo", |
|
1447 |
+ contextWindow: 131_072, maxOutputTokens: nil, |
|
1448 |
+ capabilities: ModelCapabilities(tools: true, jsonMode: true), |
|
1449 |
+ pricing: ModelPricing(inputPerMTok: 0.10, outputPerMTok: 0.32), |
|
1450 |
+ parameterSupport: .openAIDefault |
|
1451 |
+ ), |
|
1452 |
+ AIModel( |
|
1453 |
+ id: "meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo", provider: .deepinfra, displayName: "Llama 3.1 8B Turbo", |
|
1454 |
+ contextWindow: 131_072, maxOutputTokens: nil, |
|
1455 |
+ capabilities: ModelCapabilities(tools: true, jsonMode: true), |
|
1456 |
+ pricing: ModelPricing(inputPerMTok: 0.02, outputPerMTok: 0.04), |
|
1457 |
+ parameterSupport: .openAIDefault |
|
1458 |
+ ), |
|
1459 |
+ AIModel( |
|
1460 |
+ id: "openai/gpt-oss-120b", provider: .deepinfra, displayName: "GPT-OSS 120B", |
|
1461 |
+ contextWindow: 131_072, maxOutputTokens: nil, |
|
1462 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1463 |
+ pricing: ModelPricing(inputPerMTok: 0.037, outputPerMTok: 0.17), |
|
1464 |
+ parameterSupport: ParameterSupport(frequencyPenalty: true, presencePenalty: true, reasoningEffort: true), |
|
1465 |
+ isRecommended: true |
|
1466 |
+ ), |
|
1467 |
+ AIModel( |
|
1468 |
+ id: "openai/gpt-oss-20b", provider: .deepinfra, displayName: "GPT-OSS 20B", |
|
1469 |
+ contextWindow: 131_072, maxOutputTokens: nil, |
|
1470 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1471 |
+ pricing: ModelPricing(inputPerMTok: 0.03, outputPerMTok: 0.14), |
|
1472 |
+ parameterSupport: ParameterSupport(frequencyPenalty: true, presencePenalty: true, reasoningEffort: true) |
|
1473 |
+ ), |
|
1474 |
+ AIModel( |
|
1475 |
+ id: "MiniMaxAI/MiniMax-M3", provider: .deepinfra, displayName: "MiniMax M3", |
|
1476 |
+ contextWindow: 524_288, maxOutputTokens: nil, |
|
1477 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1478 |
+ pricing: ModelPricing(inputPerMTok: 0.30, outputPerMTok: 1.20), |
|
1479 |
+ parameterSupport: .openAIDefault |
|
1480 |
+ ), |
|
1481 |
+ AIModel( |
|
1482 |
+ id: "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B", provider: .deepinfra, displayName: "Nemotron 3 Ultra 550B", |
|
1483 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
1484 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1485 |
+ pricing: ModelPricing(inputPerMTok: 0.50, outputPerMTok: 2.20), |
|
1486 |
+ parameterSupport: .openAIDefault |
|
1487 |
+ ), |
|
1488 |
+ AIModel( |
|
1489 |
+ id: "mistralai/Mistral-Small-3.2-24B-Instruct-2506", provider: .deepinfra, displayName: "Mistral Small 3.2 24B", |
|
1490 |
+ contextWindow: 128_000, maxOutputTokens: nil, |
|
1491 |
+ capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
|
1492 |
+ pricing: ModelPricing(inputPerMTok: 0.075, outputPerMTok: 0.20), |
|
1493 |
+ parameterSupport: .openAIDefault |
|
1494 |
+ ), |
|
1495 |
+ AIModel( |
|
1496 |
+ id: "google/gemma-4-31B-it", provider: .deepinfra, displayName: "Gemma 4 31B", |
|
1497 |
+ contextWindow: 262_144, maxOutputTokens: nil, |
|
1498 |
+ capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true), |
|
1499 |
+ pricing: ModelPricing(inputPerMTok: 0.13, outputPerMTok: 0.38), |
|
1500 |
+ parameterSupport: .openAIDefault |
|
1501 |
+ ), |
|
1502 |
+ ] |
|
1503 |
+ |
|
1504 |
+ // MARK: - Cerebras |
|
1505 |
+ |
|
1506 |
+ static let cerebras: [AIModel] = [ |
|
1507 |
+ AIModel( |
|
1508 |
+ id: "gpt-oss-120b", provider: .cerebras, displayName: "GPT-OSS 120B", |
|
1509 |
+ contextWindow: 131_072, maxOutputTokens: 40_000, |
|
1510 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1511 |
+ pricing: ModelPricing(inputPerMTok: 0.35, outputPerMTok: 0.75), |
|
1512 |
+ parameterSupport: ParameterSupport(frequencyPenalty: true, presencePenalty: true, usesMaxCompletionTokens: true, reasoningEffort: true), |
|
1513 |
+ isRecommended: true |
|
1514 |
+ ), |
|
1515 |
+ AIModel( |
|
1516 |
+ id: "gemma-4-31b", provider: .cerebras, displayName: "Gemma 4 31B", |
|
1517 |
+ contextWindow: 131_072, maxOutputTokens: 40_000, |
|
1518 |
+ capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true), |
|
1519 |
+ pricing: ModelPricing(inputPerMTok: 0.99, outputPerMTok: 1.49), |
|
1520 |
+ parameterSupport: ParameterSupport(frequencyPenalty: true, presencePenalty: true, usesMaxCompletionTokens: true, reasoningEffort: true) |
|
1521 |
+ ), |
|
1522 |
+ AIModel( |
|
1523 |
+ id: "zai-glm-4.7", provider: .cerebras, displayName: "GLM 4.7", |
|
1524 |
+ contextWindow: 131_072, maxOutputTokens: 40_000, |
|
1525 |
+ capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true), |
|
1526 |
+ pricing: ModelPricing(inputPerMTok: 2.25, outputPerMTok: 2.75), |
|
1527 |
+ parameterSupport: ParameterSupport(frequencyPenalty: true, presencePenalty: true, usesMaxCompletionTokens: true, reasoningEffort: true), |
|
1528 |
+ // Scheduled for discontinuation on 2026-08-17. |
|
1529 |
+ isLegacy: true |
|
1530 |
+ ), |
|
1531 |
+ ] |
|
1532 |
+ |
|
1533 |
+ // MARK: - All providers |
|
1534 |
+ |
|
1535 |
+ static let all: [AIModel] = |
|
1536 |
+ openai + anthropic + xai + mistral + gemini + qwen + |
|
1537 |
+ deepseek + kimi + perplexity + together + deepinfra + cerebras |
| 41 |
1538 |
} |