Data as of November 13, 2026 · How the score is built
Decision readingSarvam 105B scores 29.4 out of 100 and ranks #174 of 215. This profile shows 10 source-displayable benchmark rows; its strongest eligible category is Instruction Following at #127. Published weights can be self-hosted, but infrastructure cost varies and is not a comparable API token rate.
Email me when its price, rank or availability changes.
Capability
29.4/100
field median 49.4#174 of 215 ranked models
Public
#174of 215
Verified —
Price
Self-hosted; infrastructure cost varies
input median $0.60No comparable first-party hosted token rate
Speed
191tok/s
field median 92 tok/sFirst token 12.11 s
Context
128Ktokens
field median 256,000Reported for this model; direct source link not stored
Decision snapshot
Where this model sits in the ranked field, its strongest published evidence, and what to validate before choosing it. The marker compares it with the current ranked catalog, not with an absolute quality threshold.
Strongest published evidence
Instruction Following ranks #127. A well-rounded choice across a range of tasks.
Validate before choosing
10 published rows leave some tracked benchmark slots empty. No comparable first-party API token rate is published.
Source-linked · 10 displayable benchmark rows
Is Sarvam 105B right for your job?
Its place on the LLM Selector’s job rankings, the same order the job pages publish. Describe your own job to see where it stands for that.
Where Sarvam 105B ranks by benchmark score:
We keep what you type here, drafts included, for 180 days, and after that only the job and requirements read from it. Nothing is tied to your IP address or account. Leave out names and other personal details. Privacy
Category score record
Scores and ranks appear only where published evidence can be displayed. The table keeps the score, weight, cohort, and evidence state together.
| Category | Score | Rank | Percentile | Weight | Benchmarks | Evidence |
|---|---|---|---|---|---|---|
| AgenticRank Not rankedWeight 22%1 benchmarkVerified | 25.2 | 1 benchmark | Verified | |||
| CodingRank Not rankedWeight 20%1 benchmarkVerified | 25.2 | 1 benchmark | Verified | |||
| ReasoningRank #131 of 160Percentile 18thWeight 17%1 benchmarkVerified | 26.3 | 1 benchmark | Verified | |||
| MultimodalWeight 12%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| KnowledgeRank #139 of 180Percentile 23rdWeight 12%2 benchmarksVerified | 33.3 | 2 benchmarks | Verified | |||
| MultilingualWeight 7%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| Inst. FollowingRank #127 of 157Percentile 19thWeight 5%1 benchmarkVerified | 39.1 | 1 benchmark | Verified | |||
| MathWeight 5%4 benchmarksVerified | Score pending | 4 benchmarks | Verified |
10 of 919 tracked benchmark slots have displayable evidence · bars run 0–100
Coverage detailsHow much of this is verified
Coverage is split by category so a strong number never hides a thin evidence base. Verified means the row is tied to a published source; provisional rows remain visible but separate.
- Agentic1/1 verified
- Coding1/1 verified
- Reasoning1/1 verified
- MultimodalNot measured
- Knowledge2/2 verified
- MultilingualNot measured
- Inst. Following1/1 verified
- Math4/4 verified
Capability shape
Each axis shows percentile within that category’s eligible cohort. The comparison outline is the median of the six nearest public-score peers; a collapsed vertex means the category is not rank-eligible.
Sarvam 105B category percentile values
- AgenticNot eligible
- CodingNot eligible
- Reasoning18th percentile
- MultimodalNot eligible
- Knowledge23rd percentile
- MultilingualNot eligible
- Instruction following19th percentile
- MathNot eligible
The dashed outline is median of 6 nearest peers.
Eligible category ranks
- AgenticNot ranked
- CodingNot ranked
- Reasoning#131/160
- MultimodalNot ranked
- Knowledge#139/180
- MultilingualNot ranked
- Inst. Following#127/157
- MathNot ranked
Benchmark ledger
Every result says who ran the test. For this model, Sarvam ran all 10. Coding opens by default. The tick marks the best source-verified result in the catalog; provisional leaders do not set the reference.
Coding1 row
| Benchmark | Score | Versus best verified row | Gap | Who ran it | Use in ranking |
|---|---|---|---|---|---|
| SWE-bench VerifiedSoftware Engineering Benchmark Verified | 45% | Best verified: Claude Opus 5 · 96% | Gap51 behind | Reported by Sarvam | Shown only |
Agentic1 row
| Benchmark | Score | Versus best verified row | Gap | Who ran it | Use in ranking |
|---|---|---|---|---|---|
| BrowseComp | 49.5% | Best verified: Atria Dawn Preview · 92.5% | Gap43 behind | Reported by Sarvam | Counts9% ref. weight |
Reasoning1 row
| Benchmark | Score | Versus best verified row | Gap | Who ran it | Use in ranking |
|---|---|---|---|---|---|
| GPQA Diamond | 78.7% | Best verified: Sarvam 105B · 78.7% | GapHighest listed | Reported by Sarvam | Shown only |
Knowledge2 rows
| Benchmark | Score | Versus best verified row | Gap | Who ran it | Use in ranking |
|---|---|---|---|---|---|
| MMLU-ProMassive Multitask Language Understanding Professional | 81.7% | Best verified: Qwen3.7 Max · 89.6% | Gap7.9 behind | Reported by Sarvam | Counts6% ref. weight |
| MMLUMassive Multitask Language Understanding | 90.6% | Best verified: o1 · 91.8% | Gap1.2 behind | Reported by Sarvam | Shown only |
Inst. Following1 row
| Benchmark | Score | Versus best verified row | Gap | Who ran it | Use in ranking |
|---|---|---|---|---|---|
| IFEvalInstruction-Following Eval | 84.8% | Best verified: LFM2.5-24B-A2B · 96.8% | Gap12 behind | Reported by Sarvam | Shown only |
Math4 rows
| Benchmark | Score | Versus best verified row | Gap | Who ran it | Use in ranking |
|---|---|---|---|---|---|
| MATH-500MATH-500 Problem Set | 98.6% | Best verified: Ternary Bonsai 2 27B · 98.8% | Gap0.2 behind | Reported by Sarvam | Shown only |
| AIME 2025American Invitational Mathematics Examination 2025 | 88.3% | Best verified: GPT-5.2 · 100% | Gap11.7 behind | Reported by Sarvam | Shown only |
| HMMT Feb 2025Harvard-MIT Mathematics Tournament February 2025 | 85.8% | Best verified: GLM-4.7 · 97.1% | Gap11.3 behind | Reported by Sarvam | Shown only |
| HMMT Nov 2025Harvard-MIT Mathematics Tournament November 2025 | 85.8% | Best verified: Qwen3.6 Plus · 94.6% | Gap8.8 behind | Reported by Sarvam | Shown only |
Outlined bars are lab-reported. Filled bars are benchmark-run or independent. Bars run 0–100; the dark tick marks the best source-verified value.
Lineage
The sequence follows explicit supersedes links. Each score is estimated for that model; a relative can inform a sparse estimate but never sets a floor, so a newer release can score below an earlier one. Scores and prices remain blank when the corresponding public row or first-party rate is unavailable.
Base entry
Spec sheet
Each documented value carries its source. Missing fields stay visible as not sourced or not published, rather than disappearing from the page.
- API model ID
- Not published
- Context window
- 128K
- Maximum output
- Not sourced yet
- Knowledge cutoff
- Not sourced yet
- Input modalities
- textsarvam-105b model card
- Output modalities
- textsarvam-105b model card
- Parameters
- Not sourced yet
- Availability
- Not sourced yet
- Cloud regions
- Not tracked yet
- Lifecycle
- Established
- API capabilities
- Tool calling, structured outputs, and batch support are not tracked yet
- Prompt caching
- Not documented in the pricing record
- Self-host
- Sarvam 105B needs ~80GB VRAM (1× NVIDIA H100 (80GB)).
- Rate limits
- Not tracked yet
How to read this profile
The visual layer above carries the decisions. These notes preserve the model, ranking, coverage, and family context behind the numbers.
Sarvam 105B ranks #174 of 215 on the public leaderboard with a score of 29.37/100. It does not yet have enough sourced coverage for a verified position.
Sarvam 105B is a open weight model with a 128K context window. It uses an explicit reasoning mode, which can improve complex problem solving while adding latency and token use.
10 of 919 tracked benchmark slots currently have displayable evidence. Missing categories stay blank.
Its strongest eligible category is Instruction Following at #127, while its lowest eligible position is Knowledge at #139. a well-rounded choice across a range of tasks.
Last updated November 13, 2026. Runtime fields remain blank until a sourced snapshot exists.
Deployment options
Self-host and provider-specific paths stay separate from benchmark evidence so operating constraints are visible before a score becomes the whole decision.
Self-host vs API cost
Estimates at 50,000 req/day · 1000 tokens/req average.
Questions
How does Sarvam 105B perform overall in AI benchmarks?
Sarvam 105B ranks #174 out of 215 models on the public BenchAlign leaderboard, with a score of 29.37/100. Its evidence status is Estimated, and this profile shows 10 source-displayable benchmark rows. The label describes evidence depth, not a provider quality claim; inspect category rows before choosing a workload.
Is Sarvam 105B good for knowledge and understanding?
Sarvam 105B ranks #139 out of 180 eligible models for knowledge and understanding, with a public category score of 33.3/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.
Is Sarvam 105B good for coding and programming?
Sarvam 105B has source-displayable benchmark coverage for coding and programming, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
Is Sarvam 105B good for mathematics?
Sarvam 105B has source-displayable benchmark coverage for mathematics, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
Is Sarvam 105B good for reasoning and logic?
Sarvam 105B ranks #131 out of 160 eligible models for reasoning and logic, with a public category score of 26.3/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.
Is Sarvam 105B good for agentic tool use and computer tasks?
Sarvam 105B has source-displayable benchmark coverage for agentic tool use and computer tasks, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
Is Sarvam 105B good for instruction following?
Sarvam 105B ranks #127 out of 157 eligible models for instruction following, with a public category score of 39.1/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.
Is Sarvam 105B open source?
Sarvam 105B is an open-weight model from Sarvam. Its weights can be downloaded for local or hosted deployment, subject to the published license. Open weight does not automatically mean open source: training data and training code may remain private, and commercial restrictions can still apply.
Does Sarvam 105B have full benchmark coverage on BenchLM?
No. Sarvam 105B currently has 16 source-displayable rows. The profile exposes published, non-generated evidence and leaves missing categories blank until an exact evaluation is available. Coverage describes how much was measured; it is not a penalty added to an individual benchmark result.
What is the context window size of Sarvam 105B?
Sarvam 105B has a reported context window of 128K in the exact-model catalog record. The value stays visible, but the profile marks its source link as unavailable instead of presenting it as directly documented. Maximum output length remains separate because providers often publish a different limit.