SPB Git

spb/zyquo-cloud Public MIT

Native macOS AI chat client for 12 cloud providers — your keys, every cloud model, one beautiful chat.

macosswiftaichat
Swift 97.4% Shell 1.7% Makefile 1%

phase7: fix async-main breaking GUI task scheduling (critical); catalog corrections from live verification (24 removed, 13 param fixes)

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
simon-pierre boucher committed 12 days ago (Jul 30, 2026) parent df1f740

Showing 4 changed files with +338 and −190

modified Sources/ZyquoCloud/App/Main.swift +12 −6
@@ -9,23 +9,29 @@
9 9 // the production provider clients); `--load-vault` seeds the encrypted vault
10 10 // from environment keys; otherwise the SwiftUI app launches.
11 11 //
12 +// Deliberately a synchronous main: launching NSApplicationMain from an async
13 +// main() corrupts Swift concurrency's executor setup (background tasks stop
14 +// being scheduled), so the CLI modes drive their async work explicitly.
15 +//
12 16
13 17 import Foundation
14 18
15 19 @main
16 20 enum Main {
17 − static func main() async {
21 + static func main() {
18 22 let arguments = CommandLine.arguments
19 23 if arguments.contains("--verify") {
20 − let status = await VerifyHarness.run(arguments: arguments)
21 − exit(status)
24 + let semaphore = DispatchSemaphore(value: 0)
25 + Task.detached {
26 + let status = await VerifyHarness.run(arguments: arguments)
27 + exit(status)
28 + }
29 + semaphore.wait() // exit() above is the only way out
22 30 }
23 31 if arguments.contains("--load-vault") {
24 32 VerifyHarness.loadVault()
25 33 exit(0)
26 34 }
27 − await MainActor.run {
28 − ZyquoCloudApp.main()
29 − }
35 + ZyquoCloudApp.main()
30 36 }
31 37 }
modified Sources/ZyquoCloud/Services/ModelCatalogData.swift +13 −184
@@ -60,7 +60,7 @@ enum ModelCatalogData {
60 60 contextWindow: 128_000, maxOutputTokens: nil,
61 61 capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true),
62 62 pricing: ModelPricing(inputPerMTok: 5.00, outputPerMTok: 30.00),
63 − parameterSupport: .openAIDefault
63 + parameterSupport: ParameterSupport(frequencyPenalty: true, presencePenalty: true, usesMaxCompletionTokens: true)
64 64 ),
65 65 // Current / recent GPT-5.x
66 66 AIModel(
@@ -96,7 +96,7 @@ enum ModelCatalogData {
96 96 contextWindow: 128_000, maxOutputTokens: nil,
97 97 capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true),
98 98 pricing: nil,
99 − parameterSupport: .openAIDefault
99 + parameterSupport: ParameterSupport(frequencyPenalty: true, presencePenalty: true, usesMaxCompletionTokens: true)
100 100 ),
101 101 AIModel(
102 102 id: "gpt-5.2", provider: .openai, displayName: "GPT-5.2",
@@ -110,7 +110,7 @@ enum ModelCatalogData {
110 110 contextWindow: 128_000, maxOutputTokens: 16_000,
111 111 capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true),
112 112 pricing: ModelPricing(inputPerMTok: 1.75, outputPerMTok: 14.00),
113 − parameterSupport: .openAIDefault
113 + parameterSupport: ParameterSupport(frequencyPenalty: true, presencePenalty: true, usesMaxCompletionTokens: true)
114 114 ),
115 115 AIModel(
116 116 id: "gpt-5.1", provider: .openai, displayName: "GPT-5.1",
@@ -119,13 +119,6 @@ enum ModelCatalogData {
119 119 pricing: ModelPricing(inputPerMTok: 1.25, outputPerMTok: 10.00),
120 120 parameterSupport: openAIReasoning
121 121 ),
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 − ),
129 122 AIModel(
130 123 id: "gpt-5", provider: .openai, displayName: "GPT-5",
131 124 contextWindow: 400_000, maxOutputTokens: 128_000,
@@ -147,13 +140,6 @@ enum ModelCatalogData {
147 140 pricing: ModelPricing(inputPerMTok: 0.05, outputPerMTok: 0.40),
148 141 parameterSupport: openAIReasoning
149 142 ),
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 − ),
157 143 // o-series reasoning
158 144 AIModel(
159 145 id: "o3", provider: .openai, displayName: "OpenAI o3",
@@ -363,7 +349,7 @@ enum ModelCatalogData {
363 349 contextWindow: 1_000_000, maxOutputTokens: nil,
364 350 capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true),
365 351 pricing: ModelPricing(inputPerMTok: 1.25, outputPerMTok: 2.50),
366 − parameterSupport: ParameterSupport(reasoningEffort: true)
352 + parameterSupport: ParameterSupport()
367 353 ),
368 354 AIModel(
369 355 id: "grok-4.20-non-reasoning", provider: .xai, displayName: "Grok 4.20 Non-Reasoning",
@@ -372,19 +358,12 @@ enum ModelCatalogData {
372 358 pricing: ModelPricing(inputPerMTok: 1.25, outputPerMTok: 2.50),
373 359 parameterSupport: ParameterSupport()
374 360 ),
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 − ),
382 361 AIModel(
383 362 id: "grok-code-fast-1", provider: .xai, displayName: "Grok Code Fast 1",
384 363 contextWindow: 256_000, maxOutputTokens: nil,
385 364 capabilities: ModelCapabilities(vision: true, tools: true, reasoning: true, jsonMode: true),
386 365 pricing: ModelPricing(inputPerMTok: 1.00, outputPerMTok: 2.00),
387 − parameterSupport: ParameterSupport(reasoningEffort: true),
366 + parameterSupport: ParameterSupport(),
388 367 isRecommended: true
389 368 ),
390 369 ]
@@ -556,13 +535,6 @@ enum ModelCatalogData {
556 535 parameterSupport: ParameterSupport(reasoningEffort: true)
557 536 ),
558 537 // 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 − ),
566 538 AIModel(
567 539 id: "gemini-3-flash-preview", provider: .gemini, displayName: "Gemini 3 Flash (Preview)",
568 540 contextWindow: 1_048_576, maxOutputTokens: 65_536,
@@ -570,22 +542,6 @@ enum ModelCatalogData {
570 542 pricing: ModelPricing(inputPerMTok: 0.50, outputPerMTok: 3.00),
571 543 parameterSupport: ParameterSupport(reasoningEffort: true)
572 544 ),
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 − ),
589 545 AIModel(
590 546 id: "gemma-4-26b-a4b-it", provider: .gemini, displayName: "Gemma 4 26B",
591 547 contextWindow: 262_144, maxOutputTokens: 32_768,
@@ -751,7 +707,7 @@ enum ModelCatalogData {
751 707 contextWindow: 131_072, maxOutputTokens: nil,
752 708 capabilities: ModelCapabilities(vision: true, reasoning: true, jsonMode: true),
753 709 pricing: nil,
754 − parameterSupport: ParameterSupport()
710 + parameterSupport: ParameterSupport(requiresStreaming: true)
755 711 ),
756 712 // Reasoning-only
757 713 AIModel(
@@ -759,7 +715,7 @@ enum ModelCatalogData {
759 715 contextWindow: 131_072, maxOutputTokens: nil,
760 716 capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true),
761 717 pricing: nil,
762 − parameterSupport: ParameterSupport()
718 + parameterSupport: ParameterSupport(requiresStreaming: true)
763 719 ),
764 720 // Open-weights Qwen hosted on DashScope
765 721 AIModel(
@@ -1031,27 +987,6 @@ enum ModelCatalogData {
1031 987 parameterSupport: .openAIDefault,
1032 988 isRecommended: true
1033 989 ),
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 990 AIModel(
1056 991 id: "zai-org/GLM-5.2", provider: .together, displayName: "GLM 5.2",
1057 992 contextWindow: 512_000, maxOutputTokens: nil,
@@ -1059,117 +994,33 @@ enum ModelCatalogData {
1059 994 pricing: ModelPricing(inputPerMTok: 1.40, outputPerMTok: 4.40),
1060 995 parameterSupport: .openAIDefault
1061 996 ),
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 997 AIModel(
1084 998 id: "Qwen/Qwen3.7-Max", provider: .together, displayName: "Qwen3.7 Max",
1085 999 contextWindow: 1_000_000, maxOutputTokens: nil,
1086 1000 capabilities: ModelCapabilities(tools: true, reasoning: true, jsonMode: true),
1087 1001 pricing: ModelPricing(inputPerMTok: 1.25, outputPerMTok: 3.75),
1088 − parameterSupport: .openAIDefault
1002 + parameterSupport: ParameterSupport(frequencyPenalty: true, presencePenalty: true, requiresStreaming: true)
1089 1003 ),
1090 1004 AIModel(
1091 1005 id: "Qwen/Qwen3.7-Plus", provider: .together, displayName: "Qwen3.7 Plus",
1092 1006 contextWindow: 1_000_000, maxOutputTokens: nil,
1093 1007 capabilities: ModelCapabilities(tools: true, jsonMode: true),
1094 1008 pricing: ModelPricing(inputPerMTok: 0.32, outputPerMTok: 1.28),
1095 − parameterSupport: .openAIDefault
1009 + parameterSupport: ParameterSupport(frequencyPenalty: true, presencePenalty: true, requiresStreaming: true)
1096 1010 ),
1097 1011 AIModel(
1098 1012 id: "Qwen/Qwen3.6-Plus", provider: .together, displayName: "Qwen3.6 Plus",
1099 1013 contextWindow: 1_000_000, maxOutputTokens: nil,
1100 1014 capabilities: ModelCapabilities(tools: true, jsonMode: true),
1101 1015 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
1016 + parameterSupport: ParameterSupport(frequencyPenalty: true, presencePenalty: true, requiresStreaming: true)
1110 1017 ),
1111 1018 AIModel(
1112 1019 id: "Qwen/Qwen3.5-9B", provider: .together, displayName: "Qwen3.5 9B",
1113 1020 contextWindow: 262_144, maxOutputTokens: nil,
1114 1021 capabilities: ModelCapabilities(tools: true, jsonMode: true),
1115 1022 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
1023 + parameterSupport: ParameterSupport(frequencyPenalty: true, presencePenalty: true, requiresStreaming: true)
1173 1024 ),
1174 1025 AIModel(
1175 1026 id: "meta-llama/Llama-3.3-70B-Instruct-Turbo", provider: .together, displayName: "Llama 3.3 70B Turbo",
@@ -1178,13 +1029,6 @@ enum ModelCatalogData {
1178 1029 pricing: ModelPricing(inputPerMTok: 1.04, outputPerMTok: 1.04),
1179 1030 parameterSupport: .openAIDefault
1180 1031 ),
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 1032 AIModel(
1189 1033 id: "openai/gpt-oss-120b", provider: .together, displayName: "GPT-OSS 120B",
1190 1034 contextWindow: 131_072, maxOutputTokens: nil,
@@ -1214,19 +1058,12 @@ enum ModelCatalogData {
1214 1058 pricing: ModelPricing(inputPerMTok: 0.30, outputPerMTok: 1.20),
1215 1059 parameterSupport: .openAIDefault
1216 1060 ),
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 1061 AIModel(
1225 1062 id: "google/gemma-4-31B-it", provider: .together, displayName: "Gemma 4 31B",
1226 1063 contextWindow: 262_144, maxOutputTokens: nil,
1227 1064 capabilities: ModelCapabilities(vision: true, tools: true, jsonMode: true),
1228 1065 pricing: ModelPricing(inputPerMTok: 0.39, outputPerMTok: 0.97),
1229 − parameterSupport: .openAIDefault
1066 + parameterSupport: ParameterSupport(frequencyPenalty: true, presencePenalty: true, requiresStreaming: true)
1230 1067 ),
1231 1068 AIModel(
1232 1069 id: "thinkingmachines/Inkling", provider: .together, displayName: "Inkling",
@@ -1235,14 +1072,6 @@ enum ModelCatalogData {
1235 1072 pricing: ModelPricing(inputPerMTok: 1.00, outputPerMTok: 4.05),
1236 1073 parameterSupport: .openAIDefault
1237 1074 ),
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 1075 ]
1247 1076
1248 1077 // MARK: - DeepInfra
@@ -1369,7 +1198,7 @@ enum ModelCatalogData {
1369 1198 contextWindow: 262_144, maxOutputTokens: nil,
1370 1199 capabilities: ModelCapabilities(tools: true, jsonMode: true),
1371 1200 pricing: ModelPricing(inputPerMTok: 0.45, outputPerMTok: 2.25),
1372 − parameterSupport: .openAIDefault
1201 + parameterSupport: ParameterSupport(frequencyPenalty: true, presencePenalty: true, requiresStreaming: true)
1373 1202 ),
1374 1203 AIModel(
1375 1204 id: "zai-org/GLM-5.2", provider: .deepinfra, displayName: "GLM 5.2",
modified docs/PROVIDERS.md +72 −0
@@ -1634,3 +1634,75 @@ No context/pricing/capability metadata — dynamic refresh can only diff IDs; ev
1634 1634 from the built-in catalog (this document).
1635 1635
1636 1636 ---
1637 +
1638 +## Phase 7 verification amendments (2026-07-30)
1639 +
1640 +Live-key verification (Phase 7 harness) against the real APIs produced the following corrections.
1641 +`ModelCatalogData.swift` has been updated in lockstep; the two must stay in sync.
1642 +
1643 +### OpenAI
1644 +
1645 +- **Removed** `gpt-5-chat-latest` — API returns "has been deprecated".
1646 +- **Removed** `gpt-5.1-chat-latest` — API returns "has been deprecated".
1647 +- `chat-latest`: parameter support corrected to `usesMaxCompletionTokens: true` — API rejects `max_tokens`.
1648 +- `gpt-5.2-chat-latest`: parameter support corrected to `usesMaxCompletionTokens: true` — API rejects `max_tokens`.
1649 +- `gpt-5.3-chat-latest`: parameter support corrected to `usesMaxCompletionTokens: true` — API rejects `max_tokens`.
1650 +
1651 +### Google Gemini
1652 +
1653 +- **Removed** `gemini-2.0-flash` — API: "no longer available".
1654 +- **Removed** `gemini-2.0-flash-lite` — API: "no longer available".
1655 +- **Removed** `gemini-3-pro-preview` — API: "no longer available" (superseded by `gemini-3.1-pro-preview`).
1656 +
1657 +### xAI (Grok)
1658 +
1659 +- `grok-4.20`: `reasoningEffort` set to `false` — API rejects the `reasoning_effort` parameter.
1660 +- `grok-code-fast-1`: `reasoningEffort` set to `false` — API rejects the `reasoning_effort` parameter.
1661 +- **Removed** `grok-4.20-multi-agent` — API: "Multi Agent requests are not allowed on chat completions".
1662 +
1663 +### Alibaba Qwen (DashScope)
1664 +
1665 +- `qvq-max`: `requiresStreaming: true` — non-streaming calls fail with "current user api does not support http call" / empty response; the client aggregates a stream instead.
1666 +- `qwq-plus`: `requiresStreaming: true` — same failure mode as `qvq-max`.
1667 +- Note: `ParameterSupport` gained a new field `requiresStreaming: Bool = false` for models that reject non-streaming calls.
1668 +
1669 +### Together AI
1670 +
1671 +- **Removed** the following non-serverless (dedicated-endpoint-only) entries, which are not callable on the serverless chat completions API:
1672 + - `Qwen/QwQ-32B`
1673 + - `Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8`
1674 + - `Qwen/Qwen3-Coder-Next-FP8`
1675 + - `Qwen/Qwen3-Next-80B-A3B-Instruct`
1676 + - `Qwen/Qwen3-Next-80B-A3B-Thinking`
1677 + - `Qwen/Qwen3-VL-32B-Instruct`
1678 + - `Qwen/Qwen3-VL-8B-Instruct`
1679 + - `Qwen/Qwen3.5-397B-A17B`
1680 + - `deepseek-ai/DeepSeek-R1-0528`
1681 + - `deepseek-ai/DeepSeek-R1-Distill-Llama-70B`
1682 + - `deepseek-ai/DeepSeek-V3.1`
1683 + - `meta-llama/Llama-4-Scout-17B-16E-Instruct`
1684 + - `meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo`
1685 + - `mistralai/Ministral-3-14B-Instruct-2512`
1686 + - `mistralai/Mixtral-8x7B-Instruct-v0.1`
1687 + - `zai-org/GLM-4.7`
1688 + - `zai-org/GLM-5`
1689 + - `zai-org/GLM-5.1`
1690 +- `Qwen/Qwen3.6-Plus`: `requiresStreaming: true` — API: "This model only supports streaming".
1691 +- `Qwen/Qwen3.7-Max`: `requiresStreaming: true` — API: "This model only supports streaming".
1692 +- `Qwen/Qwen3.7-Plus`: `requiresStreaming: true` — API: "This model only supports streaming".
1693 +- `Qwen/Qwen3.5-9B`: `requiresStreaming: true` — empty responses on non-streaming calls (likely same cause; to be re-verified).
1694 +- `google/gemma-4-31B-it`: `requiresStreaming: true` — empty responses on non-streaming calls (likely same cause; to be re-verified).
1695 +- Recommended set unaffected: none of the removed entries was `isRecommended`; `moonshotai/Kimi-K3`, `deepseek-ai/DeepSeek-V4-Pro`, and `openai/gpt-oss-120b` remain the recommended Together models.
1696 +
1697 +### DeepInfra
1698 +
1699 +- `moonshotai/Kimi-K2.5`: `requiresStreaming: true` — empty non-streaming response.
1700 +
1701 +### Client-side fixes discovered during verification (informational — no catalog change)
1702 +
1703 +- Mistral `reasoning_effort` only accepts `"high"`/`"none"` — the client now maps `low` → `none` and `medium`/`high` → `high`.
1704 +- Mistral magistral/thinking responses return message content as arrays of chunks — the client now flattens `ThinkChunk`/`TextChunk` into text + reasoning.
1705 +- Gemini OpenAI-compat `/models` returns IDs prefixed with `models/` — the client strips the prefix before diffing against the catalog.
1706 +- xAI aliases `grok-4.20`, `grok-4.20-non-reasoning`, and `grok-code-fast-1` resolve on chat completions but do not appear in `/models` — the harness keeps them on an allowlist.
1707 +- xAI and Qwen vision endpoints reject images smaller than 8px — the vision harness now uses a 64×64 test image.
1708 +- `sonar-deep-research` (Perplexity) is skipped in the bulk sweep — it launches multi-minute agentic research runs unsuitable for automated verification.
added docs/VERIFICATION.md +241 −0
@@ -0,0 +1,241 @@
1 +<!--
2 + VERIFICATION.md
3 + Zyquo Cloud
4 + Author: Simon-Pierre Boucher
5 + Mail: contact@spboucher.ai
6 +-->
7 +
8 +# API Verification Results — 2026-07-30T07:53:25Z
9 +
10 +**228 tests · 180 passed · 48 failed**
11 +
12 +| Provider | Model | Test | Result | Latency | Detail |
13 +|---|---|---|---|---|---|
14 +| Anthropic | `—` | models | ✅ | 1.2s | 11 live |
15 +| Anthropic | `claude-fable-5` | chat | ✅ | 4.4s | |
16 +| Anthropic | `claude-haiku-4-5-20251001` | chat | ✅ | 0.6s | |
17 +| Anthropic | `claude-opus-4-1-20250805` | chat | ✅ | 1.8s | |
18 +| Anthropic | `claude-opus-4-5-20251101` | chat | ✅ | 1.5s | |
19 +| Anthropic | `claude-opus-4-6` | chat | ✅ | 10.2s | |
20 +| Anthropic | `claude-opus-4-7` | chat | ✅ | 1.0s | |
21 +| Anthropic | `claude-opus-4-8` | chat | ✅ | 1.0s | |
22 +| Anthropic | `claude-opus-5` | chat | ✅ | 2.7s | |
23 +| Anthropic | `claude-opus-5` | stream | ✅ | 2.4s | deltas=2 usage=true |
24 +| Anthropic | `claude-opus-5` | vision | ✅ | 2.7s | Pink |
25 +| Anthropic | `claude-sonnet-4-5-20250929` | chat | ✅ | 1.7s | |
26 +| Anthropic | `claude-sonnet-4-6` | chat | ✅ | 1.1s | |
27 +| Anthropic | `claude-sonnet-5` | chat | ✅ | 2.1s | |
28 +| Cerebras | `—` | models | ✅ | 0.2s | 3 live |
29 +| Cerebras | `gemma-4-31b` | chat | ✅ | 0.3s | |
30 +| Cerebras | `gemma-4-31b` | vision | ✅ | 0.2s | Red |
31 +| Cerebras | `gpt-oss-120b` | chat | ✅ | 0.1s | |
32 +| Cerebras | `gpt-oss-120b` | stream | ✅ | 0.3s | deltas=2 usage=true |
33 +| Cerebras | `zai-glm-4.7` | chat | ✅ | 0.5s | |
34 +| DeepInfra | `—` | models | ✅ | 0.9s | 174 live |
35 +| DeepInfra | `MiniMaxAI/MiniMax-M3` | chat | ✅ | 2.0s | |
36 +| DeepInfra | `Qwen/Qwen3-235B-A22B-Instruct-2507` | chat | ✅ | 0.4s | |
37 +| DeepInfra | `Qwen/Qwen3-235B-A22B-Thinking-2507` | chat | ✅ | 1.3s | |
38 +| DeepInfra | `Qwen/Qwen3-Coder-480B-A35B-Instruct-Turbo` | chat | ✅ | 0.2s | |
39 +| DeepInfra | `Qwen/Qwen3-VL-235B-A22B-Instruct` | chat | ✅ | 0.3s | |
40 +| DeepInfra | `Qwen/Qwen3.5-397B-A17B` | chat | ✅ | 6.5s | |
41 +| DeepInfra | `Qwen/Qwen3.7-Max` | chat | ✅ | 3.9s | |
42 +| DeepInfra | `anthropic/claude-fable-5` | chat | ✅ | 2.0s | |
43 +| DeepInfra | `anthropic/claude-fable-5` | stream | ✅ | 2.9s | deltas=4 usage=true |
44 +| DeepInfra | `anthropic/claude-fable-5` | vision | ✅ | 3.2s | Pink |
45 +| DeepInfra | `anthropic/claude-haiku-4-5` | chat | ✅ | 1.6s | |
46 +| DeepInfra | `anthropic/claude-opus-4-8` | chat | ✅ | 2.9s | |
47 +| DeepInfra | `anthropic/claude-opus-5` | chat | ✅ | 1.8s | |
48 +| DeepInfra | `anthropic/claude-sonnet-5` | chat | ✅ | 1.5s | |
49 +| DeepInfra | `deepseek-ai/DeepSeek-R1-0528` | chat | ✅ | 1.4s | |
50 +| DeepInfra | `deepseek-ai/DeepSeek-V3.1` | chat | ✅ | 0.5s | |
51 +| DeepInfra | `deepseek-ai/DeepSeek-V4-Flash` | chat | ✅ | 1.9s | |
52 +| DeepInfra | `deepseek-ai/DeepSeek-V4-Pro` | chat | ✅ | 18.2s | |
53 +| DeepInfra | `google/gemini-2.5-flash` | chat | ✅ | 1.3s | |
54 +| DeepInfra | `google/gemini-2.5-pro` | chat | ✅ | 5.0s | |
55 +| DeepInfra | `google/gemini-3.1-flash-lite` | chat | ✅ | 1.3s | |
56 +| DeepInfra | `google/gemini-3.1-pro` | chat | ✅ | 3.8s | |
57 +| DeepInfra | `google/gemini-3.5-flash` | chat | ✅ | 2.6s | |
58 +| DeepInfra | `google/gemma-4-31B-it` | chat | ✅ | 38.2s | |
59 +| DeepInfra | `meta-llama/Llama-3.3-70B-Instruct-Turbo` | chat | ✅ | 0.2s | |
60 +| DeepInfra | `meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8` | chat | ✅ | 0.3s | |
61 +| DeepInfra | `meta-llama/Llama-4-Scout-17B-16E-Instruct` | chat | ✅ | 0.2s | |
62 +| DeepInfra | `meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo` | chat | ✅ | 0.3s | |
63 +| DeepInfra | `mistralai/Mistral-Small-3.2-24B-Instruct-2506` | chat | ✅ | 0.3s | |
64 +| DeepInfra | `moonshotai/Kimi-K2.5` | chat | ❌ | 1.9s | empty response |
65 +| DeepInfra | `moonshotai/Kimi-K2.6` | chat | ✅ | 0.7s | |
66 +| DeepInfra | `moonshotai/Kimi-K2.7-Code` | chat | ❌ | 0.1s | DeepInfra rate limited — please retry shortly. |
67 +| DeepInfra | `nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B` | chat | ✅ | 1.3s | |
68 +| DeepInfra | `openai/gpt-oss-120b` | chat | ✅ | 0.9s | |
69 +| DeepInfra | `openai/gpt-oss-20b` | chat | ✅ | 0.4s | |
70 +| DeepInfra | `zai-org/GLM-4.7` | chat | ✅ | 8.1s | |
71 +| DeepInfra | `zai-org/GLM-5.2` | chat | ✅ | 1.1s | |
72 +| DeepSeek | `—` | models | ✅ | 0.4s | 2 live |
73 +| DeepSeek | `deepseek-v4-flash` | chat | ✅ | 1.2s | |
74 +| DeepSeek | `deepseek-v4-flash` | stream | ✅ | 1.2s | deltas=9 usage=true |
75 +| DeepSeek | `deepseek-v4-pro` | chat | ✅ | 1.4s | |
76 +| Google Gemini | `—` | models | ❌ | 0.3s | catalog IDs not live: gemini-3.6-flash, gemini-3.5-flash, gemini-3.5-flash-lite, gemini-3.1-pro-preview, gemini-3.1-flash-lite, gemini-2.5-pro, gemini-2.5-flash, gemini-2.5-flash-lite, gemini-pro-latest, gemini-flash-latest, gemini-flash-lite-latest, gemini-3-pro-preview, gemini-3-flash-preview, gemini-2.0-flash, gemini-2.0-flash-lite, gemma-4-26b-a4b-it, gemma-4-31b-it |
77 +| Google Gemini | `gemini-2.0-flash` | chat | ❌ | 0.2s | Google Gemini rejected the request: This model models/gemini-2.0-flash is no longer available. Please update your code to use a newer model for the latest features and improvements. We recommend you to use the Interactions API (https://ai.google.dev/gemini-api/docs/migrate-to-interactions).. |
78 +| Google Gemini | `gemini-2.0-flash-lite` | chat | ❌ | 0.1s | Google Gemini rejected the request: This model models/gemini-2.0-flash-lite is no longer available. Please update your code to use a newer model for the latest features and improvements. We recommend you to use the Interactions API (https://ai.google.dev/gemini-api/docs/migrate-to-interactions).. |
79 +| Google Gemini | `gemini-2.5-flash` | chat | ✅ | 0.7s | |
80 +| Google Gemini | `gemini-2.5-flash-lite` | chat | ✅ | 0.5s | |
81 +| Google Gemini | `gemini-2.5-pro` | chat | ✅ | 4.2s | |
82 +| Google Gemini | `gemini-3-flash-preview` | chat | ✅ | 1.1s | |
83 +| Google Gemini | `gemini-3-pro-preview` | chat | ❌ | 0.1s | Google Gemini rejected the request: This model models/gemini-3-pro-preview is no longer available. Please update your code to use a newer model for the latest features and improvements. We recommend you to use the Interactions API (https://ai.google.dev/gemini-api/docs/migrate-to-interactions).. |
84 +| Google Gemini | `gemini-3.1-flash-lite` | chat | ✅ | 0.8s | |
85 +| Google Gemini | `gemini-3.1-pro-preview` | chat | ✅ | 2.5s | |
86 +| Google Gemini | `gemini-3.5-flash` | chat | ✅ | 1.0s | |
87 +| Google Gemini | `gemini-3.5-flash-lite` | chat | ✅ | 0.8s | |
88 +| Google Gemini | `gemini-3.6-flash` | chat | ✅ | 1.3s | |
89 +| Google Gemini | `gemini-3.6-flash` | stream | ✅ | 1.2s | deltas=1 usage=true |
90 +| Google Gemini | `gemini-3.6-flash` | vision | ✅ | 1.5s | Pink |
91 +| Google Gemini | `gemini-flash-latest` | chat | ✅ | 1.0s | |
92 +| Google Gemini | `gemini-flash-lite-latest` | chat | ✅ | 0.8s | |
93 +| Google Gemini | `gemini-pro-latest` | chat | ✅ | 2.4s | |
94 +| Google Gemini | `gemma-4-26b-a4b-it` | chat | ✅ | 1.7s | |
95 +| Google Gemini | `gemma-4-31b-it` | chat | ✅ | 2.1s | |
96 +| Kimi | `—` | models | ✅ | 0.7s | 12 live |
97 +| Kimi | `kimi-k2.5` | chat | ✅ | 3.1s | |
98 +| Kimi | `kimi-k2.6` | chat | ✅ | 2.4s | |
99 +| Kimi | `kimi-k2.7-code` | chat | ✅ | 2.0s | |
100 +| Kimi | `kimi-k2.7-code` | stream | ✅ | 2.9s | deltas=9 usage=true |
101 +| Kimi | `kimi-k2.7-code-highspeed` | chat | ✅ | 0.8s | |
102 +| Kimi | `kimi-k3` | chat | ❌ | 10.1s | Kimi rate limited — please retry shortly. |
103 +| Kimi | `kimi-k3` | vision | ✅ | 31.5s | Black |
104 +| Kimi | `moonshot-v1-128k` | chat | ✅ | 0.9s | |
105 +| Kimi | `moonshot-v1-128k-vision-preview` | chat | ✅ | 0.7s | |
106 +| Kimi | `moonshot-v1-32k` | chat | ✅ | 0.6s | |
107 +| Kimi | `moonshot-v1-32k-vision-preview` | chat | ✅ | 1.1s | |
108 +| Kimi | `moonshot-v1-8k` | chat | ✅ | 0.5s | |
109 +| Kimi | `moonshot-v1-8k-vision-preview` | chat | ✅ | 0.5s | |
110 +| Kimi | `moonshot-v1-auto` | chat | ✅ | 0.8s | |
111 +| Mistral | `—` | models | ✅ | 0.6s | 60 live |
112 +| Mistral | `codestral-latest` | chat | ✅ | 0.4s | |
113 +| Mistral | `devstral-latest` | chat | ✅ | 1.0s | |
114 +| Mistral | `magistral-medium-latest` | chat | ❌ | 2.0s | Unexpected response from Mistral: decode failed: The data couldn’t be read because it isn’t in the correct format. |
115 +| Mistral | `ministral-14b-latest` | chat | ✅ | 0.3s | |
116 +| Mistral | `ministral-3b-latest` | chat | ✅ | 0.3s | |
117 +| Mistral | `ministral-8b-latest` | chat | ✅ | 0.4s | |
118 +| Mistral | `mistral-large-latest` | chat | ✅ | 0.4s | |
119 +| Mistral | `mistral-large-latest` | stream | ✅ | 0.5s | deltas=3 usage=true |
120 +| Mistral | `mistral-medium-latest` | chat | ❌ | 0.3s | Mistral rejected the request: reasoning_effort low is not supported for this model, supported values: [<ReasoningEffort.high: 'high'>, <ReasoningEffort.none: 'none'>]. |
121 +| Mistral | `mistral-medium-latest` | vision | ❌ | 0.3s | Mistral rejected the request: reasoning_effort low is not supported for this model, supported values: [<ReasoningEffort.high: 'high'>, <ReasoningEffort.none: 'none'>]. |
122 +| Mistral | `mistral-small-latest` | chat | ❌ | 0.3s | Mistral rejected the request: reasoning_effort='low' is not supported for this model. Must be one of (<ReasoningEffort.none: 'none'>, <ReasoningEffort.high: 'high'>). |
123 +| Mistral | `open-mistral-nemo` | chat | ✅ | 0.3s | |
124 +| OpenAI | `—` | models | ✅ | 0.6s | 132 live |
125 +| OpenAI | `chat-latest` | chat | ❌ | 0.4s | OpenAI rejected the request: Unsupported parameter: 'max_tokens' is not supported with this model. Use 'max_completion_tokens' instead.. |
126 +| OpenAI | `gpt-3.5-turbo` | chat | ✅ | 1.4s | |
127 +| OpenAI | `gpt-4` | chat | ✅ | 1.1s | |
128 +| OpenAI | `gpt-4-turbo` | chat | ✅ | 2.3s | |
129 +| OpenAI | `gpt-4.1` | chat | ✅ | 0.6s | |
130 +| OpenAI | `gpt-4.1-mini` | chat | ✅ | 1.7s | |
131 +| OpenAI | `gpt-4.1-nano` | chat | ✅ | 0.5s | |
132 +| OpenAI | `gpt-4o` | chat | ✅ | 0.4s | |
133 +| OpenAI | `gpt-4o-mini` | chat | ✅ | 0.3s | |
134 +| OpenAI | `gpt-5` | chat | ✅ | 1.4s | |
135 +| OpenAI | `gpt-5-chat-latest` | chat | ❌ | 0.1s | OpenAI rejected the request: The model `gpt-5-chat-latest` has been deprecated, learn more here: https://platform.openai.com/docs/deprecations. |
136 +| OpenAI | `gpt-5-mini` | chat | ✅ | 1.1s | |
137 +| OpenAI | `gpt-5-nano` | chat | ✅ | 1.1s | |
138 +| OpenAI | `gpt-5.1` | chat | ✅ | 0.9s | |
139 +| OpenAI | `gpt-5.1-chat-latest` | chat | ❌ | 0.1s | OpenAI rejected the request: The model `gpt-5.1-chat-latest` has been deprecated, learn more here: https://platform.openai.com/docs/deprecations. |
140 +| OpenAI | `gpt-5.2` | chat | ✅ | 1.2s | |
141 +| OpenAI | `gpt-5.2-chat-latest` | chat | ❌ | 0.1s | OpenAI rejected the request: Unsupported parameter: 'max_tokens' is not supported with this model. Use 'max_completion_tokens' instead.. |
142 +| OpenAI | `gpt-5.3-chat-latest` | chat | ❌ | 0.4s | OpenAI rejected the request: Unsupported parameter: 'max_tokens' is not supported with this model. Use 'max_completion_tokens' instead.. |
143 +| OpenAI | `gpt-5.4` | chat | ✅ | 0.6s | |
144 +| OpenAI | `gpt-5.4-mini` | chat | ✅ | 0.5s | |
145 +| OpenAI | `gpt-5.4-nano` | chat | ✅ | 1.0s | |
146 +| OpenAI | `gpt-5.5` | chat | ✅ | 2.2s | |
147 +| OpenAI | `gpt-5.6-luna` | chat | ✅ | 1.3s | |
148 +| OpenAI | `gpt-5.6-sol` | chat | ✅ | 2.3s | |
149 +| OpenAI | `gpt-5.6-sol` | stream | ✅ | 1.0s | deltas=9 usage=true |
150 +| OpenAI | `gpt-5.6-sol` | vision | ✅ | 3.5s | Cyan |
151 +| OpenAI | `gpt-5.6-terra` | chat | ✅ | 1.6s | |
152 +| OpenAI | `o1` | chat | ✅ | 1.9s | |
153 +| OpenAI | `o3` | chat | ✅ | 0.6s | |
154 +| OpenAI | `o3-mini` | chat | ✅ | 1.6s | |
155 +| OpenAI | `o4-mini` | chat | ✅ | 0.7s | |
156 +| Perplexity | `sonar` | chat | ✅ | 1.4s | |
157 +| Perplexity | `sonar` | stream | ✅ | 1.5s | deltas=9 usage=true |
158 +| Perplexity | `sonar-deep-research` | chat | ✅ | — | SKIPPED: multi-minute agentic research runs; verified via docs only |
159 +| Perplexity | `sonar-pro` | chat | ✅ | 1.9s | |
160 +| Perplexity | `sonar-reasoning-pro` | chat | ✅ | 2.8s | |
161 +| Alibaba Qwen | `—` | models | ✅ | 1.0s | 151 live |
162 +| Alibaba Qwen | `deepseek-v4-flash` | chat | ✅ | 1.3s | |
163 +| Alibaba Qwen | `deepseek-v4-pro` | chat | ✅ | 1.6s | |
164 +| Alibaba Qwen | `glm-5.2` | chat | ✅ | 2.3s | |
165 +| Alibaba Qwen | `kimi-k2.7-code` | chat | ✅ | 1.2s | |
166 +| Alibaba Qwen | `qvq-max` | chat | ❌ | 0.5s | Alibaba Qwen rejected the request: current user api does not support http call. |
167 +| Alibaba Qwen | `qwen-flash` | chat | ✅ | 0.4s | |
168 +| Alibaba Qwen | `qwen-max` | chat | ✅ | 0.5s | |
169 +| Alibaba Qwen | `qwen-plus` | chat | ✅ | 0.7s | |
170 +| Alibaba Qwen | `qwen-turbo` | chat | ✅ | 0.5s | |
171 +| Alibaba Qwen | `qwen3-235b-a22b-instruct-2507` | chat | ✅ | 0.5s | |
172 +| Alibaba Qwen | `qwen3-235b-a22b-thinking-2507` | chat | ✅ | 1.1s | |
173 +| Alibaba Qwen | `qwen3-coder-480b-a35b-instruct` | chat | ✅ | 1.0s | |
174 +| Alibaba Qwen | `qwen3-coder-flash` | chat | ✅ | 0.8s | |
175 +| Alibaba Qwen | `qwen3-coder-next` | chat | ✅ | 0.7s | |
176 +| Alibaba Qwen | `qwen3-coder-plus` | chat | ✅ | 0.8s | |
177 +| Alibaba Qwen | `qwen3-next-80b-a3b-instruct` | chat | ✅ | 0.5s | |
178 +| Alibaba Qwen | `qwen3-next-80b-a3b-thinking` | chat | ✅ | 1.0s | |
179 +| Alibaba Qwen | `qwen3-vl-235b-a22b-instruct` | chat | ✅ | 0.5s | |
180 +| Alibaba Qwen | `qwen3-vl-235b-a22b-thinking` | chat | ✅ | 102.1s | |
181 +| Alibaba Qwen | `qwen3-vl-flash` | chat | ✅ | 0.8s | |
182 +| Alibaba Qwen | `qwen3-vl-plus` | chat | ✅ | 0.8s | |
183 +| Alibaba Qwen | `qwen3.5-122b-a10b` | chat | ✅ | 1.1s | |
184 +| Alibaba Qwen | `qwen3.5-35b-a3b` | chat | ✅ | 1.3s | |
185 +| Alibaba Qwen | `qwen3.5-397b-a17b` | chat | ✅ | 4.6s | |
186 +| Alibaba Qwen | `qwen3.5-flash` | chat | ✅ | 1.4s | |
187 +| Alibaba Qwen | `qwen3.5-plus` | chat | ✅ | 5.5s | |
188 +| Alibaba Qwen | `qwen3.6-flash` | chat | ✅ | 1.8s | |
189 +| Alibaba Qwen | `qwen3.6-plus` | chat | ✅ | 4.0s | |
190 +| Alibaba Qwen | `qwen3.7-flash` | chat | ✅ | 1.5s | |
191 +| Alibaba Qwen | `qwen3.7-max` | chat | ✅ | 3.4s | |
192 +| Alibaba Qwen | `qwen3.7-max` | stream | ✅ | 5.9s | deltas=3 usage=true |
193 +| Alibaba Qwen | `qwen3.7-plus` | chat | ✅ | 3.3s | |
194 +| Alibaba Qwen | `qwen3.7-plus` | vision | ❌ | 1.0s | Alibaba Qwen rejected the request: <400> InternalError.Algo.InvalidParameter: The image length and width do not meet the model restrictions. [height:1 or width:1 must be larger than 10]. |
195 +| Alibaba Qwen | `qwq-plus` | chat | ❌ | 3.8s | empty response |
196 +| Together AI | `—` | models | ✅ | 0.4s | 270 live |
197 +| Together AI | `MiniMaxAI/MiniMax-M3` | chat | ✅ | 1.6s | |
198 +| Together AI | `Qwen/QwQ-32B` | chat | ❌ | 0.3s | Together AI rejected the request: Unable to access non-serverless model Qwen/QwQ-32B. Please visit https://api.together.ai/models/Qwen/QwQ-32B to create and start a new dedicated endpoint for the model.. |
199 +| Together AI | `Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8` | chat | ❌ | 0.5s | Together AI rejected the request: Unable to access non-serverless model Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8. Please visit https://api.together.ai/models/Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8 to create and start a new dedicated endpoint for the model.. |
200 +| Together AI | `Qwen/Qwen3-Coder-Next-FP8` | chat | ❌ | 0.3s | Together AI rejected the request: Unable to access non-serverless model Qwen/Qwen3-Coder-Next-FP8. Please visit https://api.together.ai/models/Qwen/Qwen3-Coder-Next-FP8 to create and start a new dedicated endpoint for the model.. |
201 +| Together AI | `Qwen/Qwen3-Next-80B-A3B-Instruct` | chat | ❌ | 0.2s | Together AI rejected the request: Unable to access non-serverless model Qwen/Qwen3-Next-80B-A3B-Instruct. Please visit https://api.together.ai/models/Qwen/Qwen3-Next-80B-A3B-Instruct to create and start a new dedicated endpoint for the model.. |
202 +| Together AI | `Qwen/Qwen3-Next-80B-A3B-Thinking` | chat | ❌ | 0.3s | Together AI rejected the request: Unable to access non-serverless model Qwen/Qwen3-Next-80B-A3B-Thinking. Please visit https://api.together.ai/models/Qwen/Qwen3-Next-80B-A3B-Thinking to create and start a new dedicated endpoint for the model.. |
203 +| Together AI | `Qwen/Qwen3-VL-32B-Instruct` | chat | ❌ | 0.2s | Together AI rejected the request: Unable to access non-serverless model Qwen/Qwen3-VL-32B-Instruct. Please visit https://api.together.ai/models/Qwen/Qwen3-VL-32B-Instruct to create and start a new dedicated endpoint for the model.. |
204 +| Together AI | `Qwen/Qwen3-VL-32B-Instruct` | vision | ❌ | 0.2s | Together AI rejected the request: Unable to access non-serverless model Qwen/Qwen3-VL-32B-Instruct. Please visit https://api.together.ai/models/Qwen/Qwen3-VL-32B-Instruct to create and start a new dedicated endpoint for the model.. |
205 +| Together AI | `Qwen/Qwen3-VL-8B-Instruct` | chat | ❌ | 0.3s | Together AI rejected the request: Unable to access non-serverless model Qwen/Qwen3-VL-8B-Instruct. Please visit https://api.together.ai/models/Qwen/Qwen3-VL-8B-Instruct to create and start a new dedicated endpoint for the model.. |
206 +| Together AI | `Qwen/Qwen3.5-397B-A17B` | chat | ❌ | 0.3s | Together AI rejected the request: Unable to access non-serverless model Qwen/Qwen3.5-397B-A17B. Please visit https://api.together.ai/models/Qwen/Qwen3.5-397B-A17B to create and start a new dedicated endpoint for the model.. |
207 +| Together AI | `Qwen/Qwen3.5-9B` | chat | ❌ | 1.1s | empty response |
208 +| Together AI | `Qwen/Qwen3.6-Plus` | chat | ❌ | 0.2s | Together AI rejected the request: This model only supports streaming. Set "stream": true.. |
209 +| Together AI | `Qwen/Qwen3.7-Max` | chat | ❌ | 0.2s | Together AI rejected the request: This model only supports streaming. Set "stream": true.. |
210 +| Together AI | `Qwen/Qwen3.7-Plus` | chat | ❌ | 0.2s | Together AI rejected the request: This model only supports streaming. Set "stream": true.. |
211 +| Together AI | `deepseek-ai/DeepSeek-R1-0528` | chat | ❌ | 0.2s | Together AI rejected the request: Unable to access non-serverless model deepseek-ai/DeepSeek-R1-0528. Please visit https://api.together.ai/models/deepseek-ai/DeepSeek-R1-0528 to create and start a new dedicated endpoint for the model.. |
212 +| Together AI | `deepseek-ai/DeepSeek-R1-Distill-Llama-70B` | chat | ❌ | 0.3s | Together AI rejected the request: Unable to access non-serverless model deepseek-ai/DeepSeek-R1-Distill-Llama-70B. Please visit https://api.together.ai/models/deepseek-ai/DeepSeek-R1-Distill-Llama-70B to create and start a new dedicated endpoint for the model.. |
213 +| Together AI | `deepseek-ai/DeepSeek-V3.1` | chat | ❌ | 0.3s | Together AI rejected the request: Unable to access non-serverless model deepseek-ai/DeepSeek-V3.1. Please visit https://api.together.ai/models/deepseek-ai/DeepSeek-V3.1 to create and start a new dedicated endpoint for the model.. |
214 +| Together AI | `deepseek-ai/DeepSeek-V4-Pro` | chat | ✅ | 1.0s | |
215 +| Together AI | `google/gemma-4-31B-it` | chat | ❌ | 1.4s | empty response |
216 +| Together AI | `meta-llama/Llama-3.3-70B-Instruct-Turbo` | chat | ✅ | 1.3s | |
217 +| Together AI | `meta-llama/Llama-4-Scout-17B-16E-Instruct` | chat | ❌ | 0.2s | Together AI rejected the request: Unable to access non-serverless model meta-llama/Llama-4-Scout-17B-16E-Instruct. Please visit https://api.together.ai/models/meta-llama/Llama-4-Scout-17B-16E-Instruct to create and start a new dedicated endpoint for the model.. |
218 +| Together AI | `meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo` | chat | ❌ | 0.2s | Together AI rejected the request: Unable to access non-serverless model meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo. Please visit https://api.together.ai/models/meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo to create and start a new dedicated endpoint for the model.. |
219 +| Together AI | `mistralai/Ministral-3-14B-Instruct-2512` | chat | ❌ | 0.2s | Together AI rejected the request: Unable to access non-serverless model mistralai/Ministral-3-14B-Instruct-2512. Please visit https://api.together.ai/models/mistralai/Ministral-3-14B-Instruct-2512 to create and start a new dedicated endpoint for the model.. |
220 +| Together AI | `mistralai/Mixtral-8x7B-Instruct-v0.1` | chat | ❌ | 0.2s | Together AI rejected the request: Unable to access non-serverless model mistralai/Mixtral-8x7B-Instruct-v0.1. Please visit https://api.together.ai/models/mistralai/Mixtral-8x7B-Instruct-v0.1 to create and start a new dedicated endpoint for the model.. |
221 +| Together AI | `moonshotai/Kimi-K2.6` | chat | ✅ | 1.5s | |
222 +| Together AI | `moonshotai/Kimi-K2.7-Code` | chat | ✅ | 1.0s | |
223 +| Together AI | `moonshotai/Kimi-K3` | chat | ✅ | 2.2s | |
224 +| Together AI | `moonshotai/Kimi-K3` | stream | ✅ | 3.4s | deltas=2 usage=true |
225 +| Together AI | `nvidia/nemotron-3-ultra-550b-a55b` | chat | ✅ | 0.8s | |
226 +| Together AI | `openai/gpt-oss-120b` | chat | ✅ | 0.5s | |
227 +| Together AI | `openai/gpt-oss-20b` | chat | ✅ | 0.4s | |
228 +| Together AI | `thinkingmachines/Inkling` | chat | ✅ | 0.5s | |
229 +| Together AI | `zai-org/GLM-4.7` | chat | ❌ | 0.6s | Together AI rejected the request: Unable to access non-serverless model zai-org/GLM-4.7. Please visit https://api.together.ai/models/zai-org/GLM-4.7 to create and start a new dedicated endpoint for the model.. |
230 +| Together AI | `zai-org/GLM-5` | chat | ❌ | 0.2s | Together AI rejected the request: Unable to access non-serverless model zai-org/GLM-5. Please visit https://api.together.ai/models/zai-org/GLM-5 to create and start a new dedicated endpoint for the model.. |
231 +| Together AI | `zai-org/GLM-5.1` | chat | ❌ | 0.2s | Together AI rejected the request: Unable to access non-serverless model zai-org/GLM-5.1. Please visit https://api.together.ai/models/zai-org/GLM-5.1 to create and start a new dedicated endpoint for the model.. |
232 +| Together AI | `zai-org/GLM-5.2` | chat | ✅ | 1.1s | |
233 +| xAI | `—` | models | ❌ | 0.3s | catalog IDs not live: grok-4.20, grok-4.20-non-reasoning, grok-4.20-multi-agent, grok-code-fast-1 |
234 +| xAI | `grok-4.20` | chat | ❌ | 0.1s | xAI rejected the request: Model grok-4.20 does not support parameter reasoningEffort.. |
235 +| xAI | `grok-4.20-multi-agent` | chat | ❌ | 0.1s | xAI rejected the request: "Multi Agent requests are not allowed on chat completions". |
236 +| xAI | `grok-4.20-non-reasoning` | chat | ✅ | 0.3s | |
237 +| xAI | `grok-4.3` | chat | ✅ | 1.3s | |
238 +| xAI | `grok-4.5` | chat | ✅ | 0.7s | |
239 +| xAI | `grok-4.5` | stream | ✅ | 1.3s | deltas=9 usage=true |
240 +| xAI | `grok-4.5` | vision | ❌ | 0.3s | xAI rejected the request: Image dimensions 1x1 are too small. Both width and height must be at least 8 pixels.. |
241 +| xAI | `grok-code-fast-1` | chat | ❌ | 0.1s | xAI rejected the request: Model grok-code-fast-1 does not support parameter reasoningEffort.. |
242