Context
8K
Max output
32K
Serving providers
1
Cheapest input
Free tier only
On-demand listed rates, OpenRouter routed quotes and Vercel reseller quotes. Enterprise commitments (committed-use discounts, provisioned throughput, negotiated rates) can be materially lower.
Benchmarks
independent evaluations · source: Artificial AnalysisHeadline indices
Individual benchmarks · click a row for effort variants
| Benchmark | Best score | Effort | Cost / task | |
|---|---|---|---|---|
| CritPt | 0% | reasoning: true | — | |
| EvalComputeProxyFrozen Aug 18, 2026 | 152.2 | reasoning: true | — | |
| GPQA Diamond | 42% | reasoning: true | — | |
| Humanity's Last Exam | 3% | reasoning: true | — | |
| IFBench | 32% | reasoning: true | — | |
| LiveCodeBench | 30% | reasoning: true | — | |
| Long-Context Reasoning | 0% | reasoning: true | — | |
| MMLU-ProFrozen Aug 18, 2026 | 70% | reasoning: true | — | |
| Omniscience | -61.6 | reasoning: true | — | |
| Omniscience Accuracy | 0.2 | reasoning: true | — | |
| Omniscience Non Hallucination | 0.1 | reasoning: true | — | |
| SciCodeArtificial Analysis did not report this score in its latest run on Oct 14, 2026 | 18% | — | — | |
| TerminalBench Hard | 2% | reasoning: true | — | |
| τ²-bench | 0% | reasoning: true | — |
Artificial Analysis last published scores for this benchmark on Aug 18, 2026. Scores here are frozen at that date, and models released since are not on it.
Pricing by serving provider
standard tier · per 1M tokens · click a provider to expand| Serving provider | Input /1M | Output /1M | Endpoints | ||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Sarvam | Free tier only | — | 1 | ||||||||||||||||
Standardsarvam/sarvam-m
| |||||||||||||||||||
Price history
OpenRouter routed quoteOpenRouter's price is a routed quote. It moves with whichever upstream route OpenRouter prefers that day, so we do not track it as a price history.
Cost calculator
estimate your monthly spend on this model$0
estimated / month
Model IDs
copy the exact identifier for your platformsarvam/sarvam-m
Frequently asked questions
Sarvam M pricing, context and availabilityWhat is the context window of Sarvam M?
Sarvam M accepts up to 8K tokens of context and can return up to 32K output tokens.
What can Sarvam M do?
Sarvam M supports reasoning.
Listed prices from LiteLLM, OpenRouter and Vercel · last checked Oct 14, 2026
Every weekday
AI moves fast. Here's your debrief.
News, analysis, tools, and more.
For people who build with AI