Model Library
SarvamReleased March 6, 2026Sarvam 105B (high)
Sarvam · Released March 6, 2026
Intelligence score
12
At a glance
- Intelligence
- 12
- Speed
- 116.8tokens/sec
- Latency
- 2.05stime to first token
- Input price
- $0.04per 1M tokens
- Output price
- $0.17per 1M tokens
- Context window
- 130Ktokens
- Parameters
- 106billions
Summary
Comparison summary
Sarvam 105B (high) scores 12 (estimated) on the Artificial Analysis Intelligence Index, placing it above average among other open weight models of similar size (median: 9). Sarvam 105B (high) generates output at 116.8 tokens per second (based on the median across providers serving the model), which is above average compared to other open weight models of similar size (median: 88.2 t/s). Sarvam 105B (high) costs $0.04 per 1M input tokens (very competitive, median: $0.40) and $0.17 per 1M output tokens (very competitive, median: $0.84), based on the median across providers serving the model.
Specifications
Technical specifications
| Model name | Sarvam 105B (high) |
|---|---|
| Provider | Sarvam |
| Release date | March 6, 2026 |
| Intelligence score | 12 |
| Speed | 116.8 tokens/sec |
| Latency (TTFT) | 2.05s |
| Input price | $0.04 / 1M tokens |
| Output price | $0.17 / 1M tokens |
| Context window | 130,000 tokens |
| Parameters | 106B |
| Reasoning | Yes |
| Input modalities | text |
| Output modalities | — |
| Open weights | Yes |
| License | Apache 2 |
| API availability | — |
| Source | Model Intelligence legacy profile: sarvam-105b |
Pricing
Token prices
Price per 1M tokens
Sarvam 105B (high), as listed in the catalog record
- Input$0.04
- Output$0.17
| Token type | Price per 1M |
|---|---|
| Input | $0.04 |
| Output | $0.17 |
Model Intelligence legacy profile: sarvam-105b
FAQ · 20
Frequently asked questions
Sarvam 105B (high) was released on March 6, 2026.
Sarvam 105B (high) was created by Sarvam.
Sarvam 105B (high) scores 12 (estimated) on the Artificial Analysis Intelligence Index, placing it above average among other open weight models of similar size (median: 9).
Sarvam 105B (high) generates output at 116.8 tokens per second (based on the median across providers serving the model), which is above average compared to other open weight models of similar size (median: 88.2 t/s).
Sarvam 105B (high) has a time to first token (TTFT) of 2.05s (based on the median across providers serving the model), which is somewhat higher than average compared to other open weight models of similar size (median: 1.52s).
Sarvam 105B (high) costs $0.04 per 1M input tokens (very competitive, median: $0.40) and $0.17 per 1M output tokens (very competitive, median: $0.84), based on the median across providers serving the model.
Sarvam 105B (high) costs $0.04 per 1M input tokens and $0.17 per 1M output tokens (based on the median across providers serving the model). For a blended rate (7:2:1 cache hit/input/output ratio), this is $0.04 per 1M tokens. Pricing may vary by provider.
Yes, Sarvam 105B (high) is a reasoning model. It uses extended thinking or chain-of-thought reasoning to work through complex problems before providing an answer.
Sarvam 105B (high) supports text input.
Sarvam 105B (high) supports text output.
No, Sarvam 105B (high) does not support image input. It can only process text.
No, Sarvam 105B (high) is not multimodal. It only supports text input.
Sarvam 105B (high) has a context window of 130k tokens. This determines how much text and conversation history the model can process in a single request.
Yes, Sarvam 105B (high) is open weights. The model weights are publicly available and can be downloaded for self-hosting.
Sarvam 105B (high) has 106 billion parameters (10.3 billion active).
Sarvam 105B (high) is a Mixture of Experts (MoE) model with 106 billion total parameters, but only 10.3 billion active parameters are used during inference.
Sarvam 105B (high) is released under the Apache 2.0 license. This license allows commercial use.
Sarvam 105B (high) achieves a score of 12 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.
Yes, Sarvam 105B (high) is available via API through 1 provider.
Sarvam 105B (high) is available through 1 API provider.