Sarvam AI: Sarvam-105B
Sarvam-105B is a reasoning model from Sarvam AI released on Sep 1, 2025. It reads text and writes text, with a 131K token context window and up to 131K tokens per reply.
Show moreShow less
It costs $0.054 per million input tokens and $0.2124 per million output tokens. Cached input bills at $0.0336, 38% less than fresh input.
It supports tool calling and prompt caching, and its weights are open, so you can host it yourself.
3 providers serve it, including FastRouter and NanoGPT. The lowest input price is $0.04 per million tokens at FastRouter.
- Modalities
- In / out price
- $0.054 / $0.2124 per 1M
- Context
- 131K
- Max output
- 131K
- Released
- Sep 1, 2025
- Knowledge cutoff
- n/a
- Tool calling
- Structured output
- Reasoning
- Prompt caching
- Open weights
- Free to use
Sarvam-105B pricing
USD, as listed on Oct 15, 2026. Most bills are input heavy, so the cached input rate matters as much as the headline price.
| Input | $0.054 / 1M tokens | |
|---|---|---|
| Output | $0.2124 / 1M tokens | |
| Cached input (read) | $0.0336 / 1M tokens | 38% off input |
Sarvam-105B providers
3 providers serve it, including FastRouter and NanoGPT. The lowest input price is $0.04 per million tokens at FastRouter. Each host sets its own price and limits, so the cheapest one is not always the one with the longest context.
| Provider | Input /M | Output /M | Cache read /M | Context | Max output |
|---|---|---|---|---|---|
| Sarvam AI First party | n/a | n/a | – | 131K | 131K |
| FastRouter | $0.04 | $0.16 | – | 131K | 131K |
| NanoGPT | $0.054 | $0.2124 | $0.0336 | 131K | 4K |
What Sarvam-105B costs per month
Pick a workload close to yours or type your own volumes.
Supported request parameters
What you can set when you call Sarvam-105B through an OpenAI compatible API.
- Tools
tools - Reasoning
reasoning - Temperature
temperature
Alternatives to Sarvam-105B
Models that take the same inputs and call tools, closest in price, cheaper ones first.
- 262K context
Reads text, images and video. Tool calling, structured output, reasoning. Served by 5 providers.
by inclusionaiSep 8, 2026$0.06/M input tokens$0.18/M output tokens5 providers
- 524K context
Reads text. Tool calling, structured output, reasoning. Served by 2 providers.
by upstageSep 23, 2026$0.05/M input tokens$0.20/M output tokens2 providers
- 262K context
Reads text. Tool calling, structured output, reasoning. Served by 9 providers.
by nvidiaAug 11, 2026$0.05/M input tokens$0.20/M output tokens9 providers
- 1M context
Reads text. Tool calling, reasoning. Served by 5 providers.
by qwenNov 1, 2024$0.05/M input tokens$0.20/M output tokens5 providers
- 131K context
Reads text. Tool calling, reasoning.
by arcee-aiDec 1, 2025$0.045/M input tokens$0.15/M output tokens
- 128K context
Reads text, images and video. Tool calling, structured output, reasoning. Served by 6 providers.
by z-aiDec 8, 2025$0.0218/M input tokens$0.2184/M output tokens6 providers
More from Sarvam AI
- 66K context
Reads text. Tool calling, reasoning. Served by 2 providers.
by sarvamFeb 18, 2026$0.02/M input tokens$0.10/M output tokens2 providers
Sarvam-105B questions
How much does the Sarvam-105B API cost?
$0.054 per million input tokens and $0.2124 per million output tokens, with cached input at $0.0336. The estimator on this page turns that into a monthly bill for your own workload.
What is the context window of Sarvam-105B?
131,072 tokens of input, and it can write up to 131,072 tokens in one reply. As a rough guide, a million tokens is around 750,000 English words.
Which providers serve Sarvam-105B?
3 providers serve it, including FastRouter and NanoGPT. The lowest input price is $0.04 per million tokens at FastRouter. Prices differ by host, so the Providers table on this page lists each one with its input and output price.
Does Sarvam-105B support tool calling?
Yes. It accepts tool definitions, which is what an agent needs to call your APIs.
Is Sarvam-105B open source?
Its weights are published, so you can run it on your own hardware or a cloud of your choice. Read the license before commercial use.
When was Sarvam-105B released?
Sep 1, 2025.
Catalog data from models.dev (MIT licensed), refreshed daily, last on Oct 15, 2026. The summary is written from these facts.
Building on Sarvam-105B?
I test your real tasks on it and the two closest alternatives, then ship the agent on whichever wins on cost and accuracy.
Book Meeting