> This page is for Model APIs. > Sarvam Voice Agents (Samvaad) is a voice AI agent platform for Indian languages: build, test and deploy AI calling agents on your own phone numbers, from ₹3 a minute. Start here: https://docs.sarvam.ai/voice-agents/why-voice-agents. > For clean Markdown of any page, append `.md` to the page URL. > For a complete documentation index, see https://docs.sarvam.ai/llms.txt. For a section index, append `/llms.txt` to any section URL (for example https://docs.sarvam.ai/voice-agents/llms.txt). > For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.sarvam.ai/_mcp/server. # Qwen 3.8 27B > Qwen 3.8 27B is a 27B-parameter open-weight reasoning model for text and image understanding, with a 262,144-token context window and tool calling. Available on Provisioned Throughput. Qwen 3.8 27B is a 27B-parameter open-weight reasoning model served on Sarvam infrastructure. It accepts **text and images** and has a **262,144-token context window**. It thinks before answering by default, and you can turn thinking off for direct answers (see [Reasoning](#reasoning)). > **Note** > > Qwen 3.8 27B is available on **Provisioned Throughput**: capacity is set up for your workload and enabled for your organisation. Without access, requests return `404 not_found_error` and the model isn't listed in [`GET /v2/models`](/api-reference/chat/models-v2). [Contact us](/api/getting-started/help) to request access. See [Provisioned Throughput](/api/token-factory/models#provisioned-throughput). ## At a glance | | | | --------------------- | ---------------------------------------------------------------------------------------------------------------------------------- | | **Model ID** | `qwen38-27b` | | **Best for** | Mid-size reasoning, tool calling, and image understanding at dedicated capacity | | **Context window** | 262,144 tokens | | **Input → output** | Text and images → text | | **Reasoning** | On by default; turn it off with `reasoning_effort: "none"`. See [Reasoning](#reasoning) | | **Tool calling** | Supported, including `tool_choice: "required"`, named functions, and parallel tool calls | | **Log probabilities** | Supported on [Chat Completion V2](/api-reference/chat/chat-completions-v2) and [Responses](/api-reference/responses/create) | | **APIs** | [`POST /v2/chat/completions`](/api-reference/chat/chat-completions-v2) and [`POST /v2/responses`](/api-reference/responses/create) | | **Pricing** | ₹18 input · ₹5 cached input · ₹230 output per 1M tokens; reasoning is billed as output ([Pricing](/api/getting-started/pricing)) | ## Key capabilities #### Image understanding Send one or more base64 images alongside text in the same message. #### Thinking on or off Reasons by default; set `reasoning_effort: "none"` for direct, low-latency answers. #### Tool calling Use OpenAI-compatible `tools` and `tool_choice`, including parallel calls. #### Streaming Receive reasoning and answer deltas as server-sent events. ## Quickstart #### Python ```python from sarvamai import SarvamAI client = SarvamAI(api_subscription_key="YOUR_SARVAM_API_KEY") response = client.chat.completions_v2( model="qwen38-27b", messages=[ {"role": "user", "content": "Explain the difference between TCP and UDP in three bullet points."} ], max_tokens=2000, ) print(response.choices[0].message.content.strip()) ``` #### JavaScript ```javascript import { SarvamAIClient } from "sarvamai"; const client = new SarvamAIClient({ apiSubscriptionKey: "YOUR_SARVAM_API_KEY", }); async function main() { const response = await client.chat.completionsV2({ model: "qwen38-27b", messages: [ { role: "user", content: "Explain the difference between TCP and UDP in three bullet points.", }, ], max_tokens: 2000, }); console.log(response.choices[0].message.content.trim()); } main(); ``` #### cURL ```bash curl -X POST https://api.sarvam.ai/v2/chat/completions \ -H "api-subscription-key: $SARVAM_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "qwen38-27b", "messages": [ {"role": "user", "content": "Explain the difference between TCP and UDP in three bullet points."} ], "max_tokens": 2000 }' ``` The answer in `content` can start with blank lines (`"\n\n"`) when the model has reasoned first, so trim it before display. ## Reasoning Qwen 3.8 27B thinks by default. Read the trace from `choices[].message.reasoning_content` on chat, or from the `reasoning` output item on Responses. | Value | Effect | | -------------------------------------------------------------------------- | ------------------------------------------------------------------------- | | Omitted, or `"minimal"`, `"low"`, `"medium"`, `"high"`, `"xhigh"`, `"max"` | Thinking on. The levels are accepted but don't change how much it thinks. | | `"none"` | Turns thinking off; the model answers directly. | Set the value with `reasoning_effort` on [Chat Completion V2](/api-reference/chat/chat-completions-v2) or `reasoning.effort` on [Responses](/api-reference/responses/create). You can also turn thinking off with `chat_template_kwargs: {"enable_thinking": false}`, sent through `extra_body` in the SDKs. Reasoning tokens are billed as output tokens and count against `max_tokens` / `max_output_tokens`. ## Structured output Turn thinking off when you use structured output. With thinking on, the JSON is written into the reasoning trace and `content` comes back `null` on chat (on Responses, the output has only a `reasoning` item). ```python response = client.chat.completions_v2( model="qwen38-27b", reasoning_effort="none", response_format={"type": "json_object"}, messages=[ {"role": "user", "content": "Return JSON with keys city and country for the capital of Japan."} ], ) ``` ## Image input Send images as **base64 data URIs** in `image_url` content parts on chat, or `input_image` parts on Responses. Remote `http(s)` image URLs return `400`. Several images can go in one message; the model reads them in order. ## Supported parameters | Parameter | Chat Completion V2 | Responses | | ----------------------------------------------------- | :------------------------------------------------: | :-------------------------------------------------------------------: | | Streaming (`stream`) | | | | Image input (base64) | | | | Tool calling (`tools`, `tool_choice`) | `auto`, `none`, `required`, or named | | | Structured output (`response_format` / `text.format`) | with thinking off ([details](#structured-output)) | with thinking off | | Log probabilities (`logprobs`, `top_logprobs`) | | send `top_logprobs` with `include: ["message.output_text.logprobs"]` | | Multiple choices (`n` > 1) | `400` | N/A — Responses has no `n` | | `top_k` | | `400` | | Reasoning control | `reasoning_effort`; `"none"` turns thinking off | `reasoning.effort`; `"none"` turns thinking off | | Stateful fields (`store`, `previous_response_id`) | N/A | `400`; replay the conversation in `input` | `400` is `invalid_request_error`. ## Known limitations * Text and image input only. Video and audio parts return `400`. * On Chat Completion V2, a `developer` message returns `400 invalid_request_error`. Use a `system` message instead. Responses accepts `developer` input items. * With thinking on, structured output lands in the reasoning trace instead of `content`. Set `reasoning_effort: "none"` (see [Structured output](#structured-output)). * Reasoning and the final answer share the **`max_tokens`** / **`max_output_tokens`** budget. ## Limits and errors | Area | Limit or condition | API response | Recommended action | | -------------- | --------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------ | ------------------------------------------------------------------------- | | Access | Your organisation does not have Provisioned Throughput for this model | `404 not_found_error`; the model isn't listed in `GET /v2/models` | [Contact us](/api/getting-started/help) to request access. | | Context | Prompt tokens plus `max_tokens` exceed 262,144 | `400 invalid_request_error` | Shorten the prompt, send fewer images, or reduce `max_tokens`. | | Output | `max_tokens` exceeds 262,144 | `400 invalid_request_error` | Set `max_tokens` within the context window. | | Media URL | An image part uses an `http(s)` URL | `400 invalid_request_error` | Inline the image as a base64 data URI. | | Modality | Request includes video or audio input | `400 invalid_request_error` | Send text or image input. | | Messages | A `developer` message on Chat Completion V2 | `400 invalid_request_error`: "The 'developer' role is not supported for this model; send these instructions with the 'system' role." | Send the instructions as a `system` message. | | Choices | `n` > 1 | `400 invalid_request_error` | Send one request per completion. | | Reasoning | Reasoning consumes the full `max_tokens` budget | `200` with `finish_reason: "length"` | Raise `max_tokens` or turn thinking off. | | Authentication | API key is missing or invalid | `403 invalid_api_key_error` | Check the `api-subscription-key` header. | | Model | Model ID is unknown or unavailable | `404 not_found_error` | Confirm the ID with [`GET /v2/models`](/api-reference/chat/models-v2). | | Quota | No credits remain | `402 insufficient_quota_error` | Add credits in the [dashboard](https://indus.sarvam.ai/billing?ref=docs). | | Rate limit | Request or concurrency limit is exceeded | `429 rate_limit_exceeded_error` | Back off and retry. |