Repository Public
1---2title: Completions3url: https://platform.claude.com/docs/en/api/completions4---56# Completions78## Create a Text Completion910**POST** `/v1/complete`1112[Legacy] Create a Text Completion.1314The Text Completions API is a legacy API. We recommend using the [Messages API](https://platform.claude.com/docs/en/api/messages) going forward.1516Future models and features will not be compatible with Text Completions. See our [migration guide](https://platform.claude.com/docs/en/build-with-claude/working-with-messages) for guidance in migrating from Text Completions to Messages.1718### Headers1920- `"anthropic-beta": optional array of AnthropicBeta`2122 Optional header to specify the beta version(s) you want to use.2324 - `string`2526 - `"message-batches-2024-09-24" or "prompt-caching-2024-07-31" or "computer-use-2024-10-22" or 43 more`2728 - `"message-batches-2024-09-24"`2930 - `"prompt-caching-2024-07-31"`3132 - `"computer-use-2024-10-22"`3334 - `"computer-use-2025-01-24"`3536 - `"pdfs-2024-09-25"`3738 - `"token-counting-2024-11-01"`3940 - `"token-efficient-tools-2025-02-19"`4142 - `"output-128k-2025-02-19"`4344 - `"files-api-2025-04-14"`4546 - `"mcp-client-2025-04-04"`4748 - `"mcp-client-2025-11-20"`4950 - `"dev-full-thinking-2025-05-14"`5152 - `"interleaved-thinking-2025-05-14"`5354 - `"code-execution-2025-05-22"`5556 - `"extended-cache-ttl-2025-04-11"`5758 - `"context-1m-2025-08-07"`5960 - `"context-management-2025-06-27"`6162 - `"model-context-window-exceeded-2025-08-26"`6364 - `"skills-2025-10-02"`6566 - `"fast-mode-2026-02-01"`6768 - `"output-300k-2026-03-24"`6970 - `"user-profiles-2026-03-24"`7172 - `"user-profiles-2026-08-18"`7374 - `"user-profiles-2026-09-04"`7576 - `"advisor-tool-2026-03-01"`7778 - `"managed-agents-2026-04-01"`7980 - `"cache-diagnosis-2026-04-07"`8182 - `"dreaming-2026-04-21"`8384 - `"thinking-token-count-2026-05-13"`8586 - `"server-side-fallback-2026-06-01"`8788 - `"server-side-fallback-2026-07-01"`8990 - `"fallback-credit-2026-06-01"`9192 - `"fallback-credit-2026-07-01"`9394 - `"agent-memory-2026-07-22"`9596 - `"mid-conversation-tool-changes-2026-07-01"`9798 - `"compact-2026-01-12"`99100 - `"computer-use-2025-11-24"`101102 - `"mcp-tunnels-2026-06-22"`103104 - `"structured-outputs-2025-11-13"`105106 - `"task-budgets-2026-03-13"`107108 - `"thinking-display-updates-2026-08-18"`109110 - `"ce-user-management-2026-07-13"`111112 - `"mid-conversation-output-config-2026-07-01"`113114 - `"thinking-binding-controls-2026-08-01"`115116 - `"mid-conversation-system-clear-at-2026-08-21"`117118 - `"compact-2026-09-04"`119120- `"anthropic-workspace-id": optional string`121122### Body parameters123124- `max_tokens_to_sample: number`125126 The maximum number of tokens to generate before stopping.127128 Note that our models may stop _before_ reaching this maximum. This parameter only specifies the absolute maximum number of tokens to generate.129130 minimum: 1131132- `model: Model`133134 The model that will complete your prompt.135136 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.137138 - `"claude-fable-5-1" or "claude-mythos-5-1" or "claude-sonnet-5" or 14 more`139140 The model that will complete your prompt.141142 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.143144 - `"claude-fable-5-1"`145146 Frontier intelligence for ambitious tasks across coding, scientific discovery, and enterprise workflows147148 - `"claude-mythos-5-1"`149150 Our most capable model for cybersecurity and biology research, available through trusted access programs151152 - `"claude-sonnet-5"`153154 High-performance model for coding and agents155156 - `"claude-fable-5"`157158 Next generation of intelligence for the hardest knowledge work and coding problems159160 - `"claude-mythos-5"`161162 Most capable model for cybersecurity and biology research163164 - `"claude-opus-5"`165166 Powerful intelligence for long-running agents and coding167168 - `"claude-opus-4-8"`169170 Powerful intelligence for long-running agents and coding171172 - `"claude-opus-4-7"`173174 Powerful intelligence for long-running agents and coding175176 - `"claude-mythos-preview"`177178 New class of intelligence, strongest in coding and cybersecurity179180 - `"claude-opus-4-6"`181182 Powerful intelligence for long-running agents and coding183184 - `"claude-sonnet-4-6"`185186 Best combination of speed and intelligence187188 - `"claude-haiku-4-5"`189190 Fastest model with near-frontier intelligence191192 - `"claude-haiku-4-5-20251001"`193194 Fastest model with near-frontier intelligence195196 - `"claude-opus-4-5"`197198 Powerful intelligence for long-running agents and coding199200 - `"claude-opus-4-5-20251101"`201202 Powerful intelligence for long-running agents and coding203204 - `"claude-sonnet-4-5"`205206 High-performance model for agents and coding207208 - `"claude-sonnet-4-5-20250929"`209210 High-performance model for agents and coding211212 - `string`213214- `prompt: string`215216 The prompt that you want Claude to complete.217218 For proper response generation you will need to format your prompt using alternating `219220 Human:`and`221222 Assistant:` conversational turns. For example:223224 ```225 "226 227 Human: {userQuestion}228 229 Assistant:"230 ```231232 See [prompt validation](https://platform.claude.com/docs/en/build-with-claude/working-with-messages) and our guide to [prompt design](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/overview) for more details.233234 minLength: 1235236- `metadata: optional Metadata`237238 An object describing metadata about the request.239240 - `user_id: optional string or null`241242 An external identifier for the user who is associated with the request.243244 This should be a uuid, hash value, or other opaque identifier. Anthropic may use this id to help detect abuse. Do not include any identifying information such as name, email address, or phone number.245246 maxLength: 512247248- `stop_sequences: optional array of string`249250 Sequences that will cause the model to stop generating.251252 Our models stop on `"253254 Human:"`, and may include additional built-in stop sequences in the future. By providing the stop_sequences parameter, you may include additional strings that will cause the model to stop generating.255256- `stream: optional boolean`257258 Whether to incrementally stream the response using server-sent events.259260 See [streaming](https://platform.claude.com/docs/en/build-with-claude/streaming) for details.261262- `temperature: optional number`263264 **Deprecated**: Deprecated. Models released after Claude Opus 4.6 do not support setting temperature. A value of 1.0 will be accepted for backwards compatibility, all other values will be rejected with a 400 error.265266 Amount of randomness injected into the response.267268 Defaults to `1.0`. Ranges from `0.0` to `1.0`. Use `temperature` closer to `0.0` for analytical / multiple choice, and closer to `1.0` for creative and generative tasks.269270 Note that even with `temperature` of `0.0`, the results will not be fully deterministic.271272 maximum: 1, minimum: 0273274- `top_k: optional number`275276 **Deprecated**: Deprecated. Models released after Claude Opus 4.6 do not accept top_k; any value will be rejected with a 400 error.277278 Only sample from the top K options for each subsequent token.279280 Used to remove "long tail" low probability responses. [Learn more technical details here](https://towardsdatascience.com/how-to-sample-from-language-models-682bceb97277).281282 Recommended for advanced use cases only.283284 minimum: 0285286- `top_p: optional number`287288 **Deprecated**: Deprecated. Models released after Claude Opus 4.6 do not support setting top_p. A value >= 0.99 will be accepted for backwards compatibility, all other values will be rejected with a 400 error.289290 Use nucleus sampling.291292 In nucleus sampling, we compute the cumulative distribution over all the options for each subsequent token in decreasing probability order and cut it off once it reaches a particular probability specified by `top_p`.293294 Recommended for advanced use cases only.295296 maximum: 1, minimum: 0297298### Returns299300- `Completion object`301302 - `type: "completion"`303304 Object type.305306 For Text Completions, this is always `"completion"`.307308 default: completion309310 - `id: string`311312 Unique object identifier.313314 The format and length of IDs may change over time.315316 - `completion: string`317318 The resulting completion up to and excluding the stop sequences.319320 - `model: Model`321322 The model that will complete your prompt.323324 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.325326 - `"claude-fable-5-1" or "claude-mythos-5-1" or "claude-sonnet-5" or 14 more`327328 The model that will complete your prompt.329330 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.331332 - `"claude-fable-5-1"`333334 Frontier intelligence for ambitious tasks across coding, scientific discovery, and enterprise workflows335336 - `"claude-mythos-5-1"`337338 Our most capable model for cybersecurity and biology research, available through trusted access programs339340 - `"claude-sonnet-5"`341342 High-performance model for coding and agents343344 - `"claude-fable-5"`345346 Next generation of intelligence for the hardest knowledge work and coding problems347348 - `"claude-mythos-5"`349350 Most capable model for cybersecurity and biology research351352 - `"claude-opus-5"`353354 Powerful intelligence for long-running agents and coding355356 - `"claude-opus-4-8"`357358 Powerful intelligence for long-running agents and coding359360 - `"claude-opus-4-7"`361362 Powerful intelligence for long-running agents and coding363364 - `"claude-mythos-preview"`365366 New class of intelligence, strongest in coding and cybersecurity367368 - `"claude-opus-4-6"`369370 Powerful intelligence for long-running agents and coding371372 - `"claude-sonnet-4-6"`373374 Best combination of speed and intelligence375376 - `"claude-haiku-4-5"`377378 Fastest model with near-frontier intelligence379380 - `"claude-haiku-4-5-20251001"`381382 Fastest model with near-frontier intelligence383384 - `"claude-opus-4-5"`385386 Powerful intelligence for long-running agents and coding387388 - `"claude-opus-4-5-20251101"`389390 Powerful intelligence for long-running agents and coding391392 - `"claude-sonnet-4-5"`393394 High-performance model for agents and coding395396 - `"claude-sonnet-4-5-20250929"`397398 High-performance model for agents and coding399400 - `string`401402 - `stop_reason: string or null`403404 The reason that we stopped.405406 This may be one the following values:407408 * `"stop_sequence"`: we reached a stop sequence — either provided by you via the `stop_sequences` parameter, or a stop sequence built into the model409 * `"max_tokens"`: we exceeded `max_tokens_to_sample` or the model's maximum410411### Example412413```bash414curl https://api.anthropic.com/v1/complete \415 -H 'Content-Type: application/json' \416 -H 'anthropic-version: 2023-06-01' \417 -H "X-Api-Key: $ANTHROPIC_API_KEY" \418 --max-time 600 \419 -d '{420 "max_tokens_to_sample": 256,421 "model": "claude-2.1",422 "prompt": "\n\nHuman: Hello, world!\n\nAssistant:",423 "temperature": 1,424 "top_k": 5,425 "top_p": 0.7426 }'427```428429#### Response (200)430431```json432{433 "id": "compl_018CKm6gsux7P8yMcwZbeCPw",434 "completion": " Hello! My name is Claude.",435 "model": "claude-2.1",436 "stop_reason": "stop_sequence",437 "type": "completion"438}439```440441## Domain types442443### Completion444445- `Completion object`446447 - `type: "completion"`448449 Object type.450451 For Text Completions, this is always `"completion"`.452453 default: completion454455 - `id: string`456457 Unique object identifier.458459 The format and length of IDs may change over time.460461 - `completion: string`462463 The resulting completion up to and excluding the stop sequences.464465 - `model: Model`466467 The model that will complete your prompt.468469 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.470471 - `"claude-fable-5-1" or "claude-mythos-5-1" or "claude-sonnet-5" or 14 more`472473 The model that will complete your prompt.474475 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.476477 - `"claude-fable-5-1"`478479 Frontier intelligence for ambitious tasks across coding, scientific discovery, and enterprise workflows480481 - `"claude-mythos-5-1"`482483 Our most capable model for cybersecurity and biology research, available through trusted access programs484485 - `"claude-sonnet-5"`486487 High-performance model for coding and agents488489 - `"claude-fable-5"`490491 Next generation of intelligence for the hardest knowledge work and coding problems492493 - `"claude-mythos-5"`494495 Most capable model for cybersecurity and biology research496497 - `"claude-opus-5"`498499 Powerful intelligence for long-running agents and coding500501 - `"claude-opus-4-8"`502503 Powerful intelligence for long-running agents and coding504505 - `"claude-opus-4-7"`506507 Powerful intelligence for long-running agents and coding508509 - `"claude-mythos-preview"`510511 New class of intelligence, strongest in coding and cybersecurity512513 - `"claude-opus-4-6"`514515 Powerful intelligence for long-running agents and coding516517 - `"claude-sonnet-4-6"`518519 Best combination of speed and intelligence520521 - `"claude-haiku-4-5"`522523 Fastest model with near-frontier intelligence524525 - `"claude-haiku-4-5-20251001"`526527 Fastest model with near-frontier intelligence528529 - `"claude-opus-4-5"`530531 Powerful intelligence for long-running agents and coding532533 - `"claude-opus-4-5-20251101"`534535 Powerful intelligence for long-running agents and coding536537 - `"claude-sonnet-4-5"`538539 High-performance model for agents and coding540541 - `"claude-sonnet-4-5-20250929"`542543 High-performance model for agents and coding544545 - `string`546547 - `stop_reason: string or null`548549 The reason that we stopped.550551 This may be one the following values:552553 * `"stop_sequence"`: we reached a stop sequence — either provided by you via the `stop_sequences` parameter, or a stop sequence built into the model554 * `"max_tokens"`: we exceeded `max_tokens_to_sample` or the model's maximum555