Sarvam 105B
Flagship Indian-language reasoning model for enterprise multilingual applications.
- Capability
- —No ECI score yet
- Input
- $0.047per 1M tokens
- Output
- $0.186per 1M tokens
- Context
- 131K131K max output
What it costs
- Input
- $0.047per million tokens
- Output
- $0.186per million tokens
- Blended (3:1)
- $0.082cheaper than 93% of priced models
Typical monthly bills
| Monthly usage | Estimated cost |
|---|---|
| Side project2M input + 0.5M output tokens | $0.187 |
| Team assistant25M input + 5M output tokens | $2.11 |
| Production app250M input + 50M output tokens | $21.06 |
Median of 2 paid API providers on models.dev.
Independent benchmarks
Epoch AI has not published a Capabilities Index score for Sarvam 105B yet.
The details
- Lab
- Sarvam AI
- Released
- Sep 1, 2025
- Context window
- 131,072 tokens
- Max output
- 131,072 tokens
- Inputs
- Text
- Output
- Text
- Reasoning
- Adjustable effort · low · medium · high
- Tool calling
- Yes
- Structured output
- No
- Weights
- Open weights
- API model ID
sarvam-105bon Sarvam AI- Availability
- 3 API providerslisted on models.dev
More from Sarvam AI
About Sarvam 105B
How much does Sarvam 105B cost?
Sarvam 105B costs $0.047 input / $0.186 output per million tokens (median across 2 API providers). At a 3:1 input-to-output mix that is $0.082 per million tokens, cheaper than 93% of the 360 priced models we track.
What is the context window of Sarvam 105B?
Sarvam 105B accepts up to 131,072 tokens per request and can write up to 131,072 tokens in one response.
How good is Sarvam 105B?
Epoch AI has not published a Capabilities Index score for Sarvam 105B yet.
Is Sarvam 105B open source?
Yes. Sarvam AI publishes the weights, so it can be downloaded and self-hosted.
What inputs does Sarvam 105B support?
Sarvam 105B accepts text and replies in text. It is a reasoning model with low, medium and high effort settings, supports tool calling.
When was Sarvam 105B released?
Sarvam AI released Sarvam 105B on Sep 1, 2025.
What are the best alternatives to Sarvam 105B?
The closest current models from other labs on capability, price and release date are Seed 1.6 Flash (ByteDance Seed, $0.022 / $0.223), Apertus 8B (Swiss AI, $0.10 / $0.20), Granite-4.0-H-Small (IBM, $0.064 / $0.265) and Trinity Mini (Arcee AI, $0.045 / $0.15).