chat

Chat Completions — Sarvam AI

Live in-chat completion → sarvam-mcp (sarvam_tools_llm_complete). This skill = SDK code.

Important

Auth: api-subscription-key header — NOT Authorization: Bearer. Base URL: https://api.sarvam.ai/v1

Models

Model	Context	Best For
sarvam-105b	128K	Complex reasoning, coding, agentic workflows
sarvam-105b-conversations	32K	Real-time chat, voice agents, conversational AI

sarvam-30b is deprecated — migrate to sarvam-105b for complex tasks or sarvam-105b-conversations for real-time/voice (the model built to fill sarvam-30b's old slot). The fixed-context variants (sarvam-105b-32k, sarvam-30b-16k) are retired — base models serve their full context window directly.

Also available (beta, on /v2/chat/completions): open-weight models glm5.3 (1M context, reasoning), deepseekv4-flash (1M context, reasoning), gemma4 (131K context, text+image, no reasoning). Requires beta access on your API key. Called via client.chat.completions_v2(...) in Python — not .completions(). Still under beta and expected to GA soon — see Full Docs for details.

Quick Start (Python)

from sarvamai import SarvamAI
client = SarvamAI()

response = client.chat.completions(
    model="sarvam-105b-conversations",
    messages=[{"role": "user", "content": "भारत की राजधानी क्या है?"}]
)
print(response.choices[0].message.content)

Streaming (Python)

for chunk in client.chat.completions(
    model="sarvam-105b-conversations",
    messages=[{"role": "user", "content": "Write a poem about India"}],
    stream=True
):
    if chunk.choices and chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="", flush=True)

Quick Start (JavaScript/TypeScript)

import { SarvamAIClient } from "sarvamai";

const client = new SarvamAIClient({ apiSubscriptionKey: "YOUR_SARVAM_API_KEY" });

const response = await client.chat.completions({
    model: "sarvam-105b-conversations",
    messages: [{ role: "user", content: "भारत की राजधानी क्या है?" }]
});
console.log(response.choices[0].message.content);

OpenAI-Compatible (both languages)

from openai import OpenAI
client = OpenAI(api_key="your-key", base_url="https://api.sarvam.ai/v1")
response = client.chat.completions.create(model="sarvam-105b-conversations", messages=[...])

Gotchas

Gotcha	Detail
SDK method	Python: client.chat.completions(...), JS: client.chat.completions({...}) — no .create() in either. OpenAI SDK uses .create() as usual.
JS constructor	new SarvamAIClient({ apiSubscriptionKey: "..." }) — NOT SarvamAI(). Key is passed explicitly.
content can be None	Models produce reasoning_content before content. If max_tokens is too low, reasoning consumes the budget, finish_reason is "length", and content is None. Omit max_tokens, set 500+, or disable reasoning with reasoning_effort=None. Check reasoning_content as fallback. Verified live: this is real and a little flaky — the same short prompt at max_tokens=500 returned content=None once and content="OK" on retry, with reasoning eating 251 of the 500 tokens that time. Don't treat 500 as a guarantee; check reasoning_content/finish_reason regardless.
reasoning_effort	Thinking is on by default at "low". Values: "low"|"medium"|"high", or None to disable reasoning entirely. NOT thinking=True. Reasoning tokens count toward completion tokens and billing.
sarvam-30b now hard-errors, not just "deprecated"	Verified live: calling it returns 400 invalid_request_error — "Model 'sarvam-30b' has been deprecated. Please use one of the available models instead: sarvam-105b, sarvam-105b-conversations." It's not a soft warning; existing code on sarvam-30b is already broken. Migrate to sarvam-105b or sarvam-105b-conversations.
Open-weight models: Python-only for now	glm5.3 / gemma4 / deepseekv4-flash go through client.chat.completions_v2(...) on /v2/chat/completions in Python (verified working, all three, with a beta-enabled key). Verified live: the installed JS SDK ([email protected]) has no completionsV2 method at all (client.chat.completionsV2 is not a function) — only .completions() exists in JS right now. Don't generate JS code calling completionsV2 until the JS SDK ships it; check npm ls sarvamai / the changelog first.

Full Docs

Fetch detailed parameters, tool calling, streaming, and examples from:

• https://docs.sarvam.ai/llms.txt — comprehensive docs index
• Chat Completion Guide
• Model Specs
• Rate Limits