> 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. # Clef > Clef is a 27B multimodal decision model. Send a state, optional images, and typed questions to POST /v2/systemone and get a calibrated probability for every option. Available on Provisioned Throughput. Clef is a 27B open-weight **decision model**. Like [Sarvam Prism](/api/token-factory/models/sarvam-prism), it answers typed questions about an input on the [Decisions API](/api/token-factory/decisions-api): you send a **state** and a map of **questions**, and it returns a probability for every option of every question in one pass. It generates no text, so there is nothing to parse. Unlike Sarvam Prism, Clef also reads **images**, and it takes up to 65,536 tokens per request. > **Note** > > Clef 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** | `clef` | | **What it does** | Answers yes/no (`noul`), pick-one (`choice`) and ordered-scale (`score`) questions about your input, with a probability for every option | | **API** | [Decisions API](/api/token-factory/decisions-api) (`POST /v2/systemone`, [reference](/api-reference/decisions/systemone)), synchronous JSON, no streaming | | **Input** | Text, or any JSON object or array (rendered as JSON), plus optional base64 images | | **Input limits** | 65,536 tokens per request (state, all questions and options, and images), 1–255 options per `choice`, 1–10 levels per `score`. [Details](#limits) | | **Output** | Probabilities that sum to 1 for each question, plus the top option, a confidence value, and (for `score`) a legend | | **Repeatability** | The same request returns the same probabilities | | **Pricing** | ₹20 per 1M input tokens; no output tokens are billed ([Pricing](/api/getting-started/pricing)). Check `usage.input_tokens` for the exact count | ## Quickstart #### Python ```python from sarvamai import SarvamAI client = SarvamAI(api_subscription_key="YOUR_SARVAM_API_KEY") response = client.decisions.create( model="clef", state="My Stripe integration stopped working and all checkout payments are failing.", questions={ "urgent": { "type": "noul", "instructions": "Is this urgent and revenue-impacting?", }, "department": { "type": "choice", "instructions": "Which team should handle this?", "criteria": { "billing": "Payments or invoices", "technical": "Bugs or outages", "sales": "New business enquiries", }, }, "severity": { "type": "score", "instructions": "Rate the severity.", "criteria": ["trivial", "minor", "moderate", "major", "critical"], }, }, ) answers = response.answers if answers["urgent"].noul > 0.5: print("Escalate to", answers["department"].choice) ``` #### cURL ```bash curl -X POST https://api.sarvam.ai/v2/systemone \ -H "api-subscription-key: $SARVAM_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "clef", "state": "My Stripe integration stopped working and all checkout payments are failing.", "questions": { "urgent": {"type": "noul", "instructions": "Is this urgent and revenue-impacting?"}, "department": { "type": "choice", "instructions": "Which team should handle this?", "criteria": { "billing": "Payments or invoices", "technical": "Bugs or outages", "sales": "New business enquiries" } }, "severity": { "type": "score", "instructions": "Rate the severity.", "criteria": ["trivial", "minor", "moderate", "major", "critical"] } } }' ``` Response: ```json { "model": "clef", "answers": { "urgent": {"type": "noul", "noul": 0.9926}, "department": { "type": "choice", "choice": "billing", "confidence": 0.3099, "probabilities": {"billing": 0.5400, "technical": 0.4476, "sales": 0.0124} }, "severity": { "type": "score", "score": 3.8746, "confidence": 0.8955, "legend": {"0": "trivial", "1": "minor", "2": "moderate", "3": "major", "4": "critical"}, "probabilities": {"0": 0.0071, "1": 0.0068, "2": 0.0089, "3": 0.0587, "4": 0.9184} } }, "usage": {"input_tokens": 357, "output_tokens": 0} } ``` Probabilities are rounded here; the API returns full floats. For what each field means, see [Read the answers](/api/token-factory/decisions-api#read-the-answers). ## Questions about images Send images in `images`, as base64 data URIs. The questions are answered about the state and the images together. Remote `http(s)` image URLs return `400`. The `sarvamai` SDK's `decisions.create` doesn't have an `images` argument yet, so pass the images as an extra body field: ```python import base64 with open("receipt.jpg", "rb") as f: image = "data:image/jpeg;base64," + base64.b64encode(f.read()).decode() response = client.decisions.create( model="clef", state="Review the attached receipt.", questions={"legible": {"type": "noul", "instructions": "Is the receipt total legible?"}}, request_options={"additional_body_parameters": {"images": [image]}}, ) print(response.answers["legible"].noul) ``` In the JavaScript SDK, put `images` directly in the request object: `client.decisions.create({ model: "clef", state, questions, images: [image] })`. Its `requestOptions` can't add body fields. Over HTTP, send `images` in the request body: ```json { "model": "clef", "state": "Review the attached receipt.", "images": ["data:image/jpeg;base64,/9j/4AAQ..."], "questions": { "legible": {"type": "noul", "instructions": "Is the receipt total legible?"} } } ``` ## Question rules The question types are the same as on Sarvam Prism ([Question types](/api/token-factory/decisions-api#question-types)), with these rules on Clef: | `type` | `criteria` | Rules | | -------- | ------------------------------------------------------ | ------------------------------------------------------------------------------ | | `noul` | Optional `{"true": "...", "false": "..."}` | Needs `instructions`, or at least one of `criteria.true` and `criteria.false`. | | `choice` | Required object `{option_id: description or null}` | 1–255 options. | | `score` | Required **array** of level descriptions, lowest first | 1–10 levels, numbered from 0. An object returns `422`. | * Always send `instructions` (or `noul` criteria). The question ID is not used as the question. * Any key in a question other than `type`, `instructions`, and `criteria` returns `422`, including misspellings such as `instruction`. * `temperature` returns `422`. `max_tokens` and `stream` are accepted and ignored: Clef always answers in one JSON response. * `chat_template_kwargs` must be empty or absent; a non-empty value returns `400`. * `videos` and other unknown top-level fields are ignored. ## Limits * **65,536 tokens per request**, counting the state, every question and its options, and images. The state is counted once, however many questions you ask. * **The state is never truncated.** A request over the limit fails with `400 invalid_request_error`, and the message gives the count, such as "the prompt needs 70141 tokens (70001 for the state), but holds at most 65536". Shorten the state, or split a long document across requests. * Ask several questions about the same state in one request: they are answered together, in one pass. ## Differences from Sarvam Prism | | Clef | Sarvam Prism | | ---------------- | ---------------------------------------------- | ------------------------------------------- | | Images | Supported (`images`, base64) | Not supported | | Token limit | 65,536 per request | 32,768 per question | | Over-long state | Rejected with `400` | Rejected with `422` | | `choice` options | 1–255 | 2–255 | | `score` levels | 1–10, array only | 2–10 | | Repeatability | Identical probabilities for identical requests | Can vary slightly; the top answer is stable | | Access | Provisioned Throughput | Serverless | ## Errors | Status | When | Recommended action | | -------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ---------------------------------------------------------- | | `404 not_found_error` | Your organisation does not have Provisioned Throughput for Clef | [Contact us](/api/getting-started/help) to request access. | | `400 invalid_request_error` | The prompt is over 65,536 tokens, an image is a remote URL, or an image can't be read | Shorten the state, or send images as base64 data URIs. | | `400 invalid_request_error` | Clef is called on another endpoint, such as `/v2/chat/completions` ("served only on /v2/systemone") | Send decisions to `POST /v2/systemone`. | | `422 unprocessable_entity_error` | Missing `state` or `questions`, empty `questions`, unknown `type`, unknown key in a question, too many options or levels, a `noul` with nothing to decide, or `temperature` | Fix the request. `param` names the field at fault. | | `403 invalid_api_key_error` | API key is missing or invalid | Check the `api-subscription-key` header. | | `429 rate_limit_exceeded_error` | Request or concurrency limit is exceeded | Back off and retry. | | `5xx` | Server or capacity error | Retry with backoff. | Every error carries `message`, `code`, and `request_id`; validation errors also carry `param`.