Grok Code Fast 1 vs Sarvam 105B

Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.

Agentic
Coding
Multimodal & Grounded
Reasoning
Knowledge
Instruction Following
Multilingual
Mathematics

Grok Code Fast 1· Sarvam 105B

Quick Verdict

Pick Sarvam 105B if you want the stronger benchmark profile. Grok Code Fast 1 only becomes the better choice if coding is the priority or you need the larger 256K context window.

Sarvam 105B is clearly ahead on the aggregate, 60 to 56. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

Sarvam 105B's sharpest advantage is in knowledge, where it averages 81.7 against 49. The single biggest benchmark swing on the page is MMLU, 64% to 90.6%. Grok Code Fast 1 does hit back in coding, so the answer changes if that is the part of the workload you care about most.

Sarvam 105B is the reasoning model in the pair, while Grok Code Fast 1 is not. That usually helps on harder chain-of-thought-heavy tests, but it can also mean more latency and more token spend in real use. Grok Code Fast 1 gives you the larger context window at 256K, compared with 128K for Sarvam 105B.

Operational tradeoffs

ProviderxAISarvam
PricePricing unavailableFree*
Speed172 t/sN/A
TTFT2.81sN/A
Context256K128K

Decision framing

BenchLM keeps the benchmark table and the operator tradeoffs on the same page so a better score does not hide a materially slower, pricier, or smaller-context model.

Runtime metrics show N/A when BenchLM does not have a sourced snapshot for that exact model. The scoring rules and freshness policy are documented on the methodology page.

BenchmarkGrok Code Fast 1Sarvam 105B
AgenticGrok Code Fast 1 wins
Terminal-Bench 2.059%
BrowseComp57%49.5%
OSWorld-Verified51%
CodingGrok Code Fast 1 wins
HumanEval60%
SWE-bench Verified70.8%45%
LiveCodeBench80%
SWE-bench Pro42%
LiveCodeBench v671.7%
Multimodal & Grounded
MMMU-Pro40%
OfficeQA Pro63%
Reasoning
MuSR59%
BBH75%
LongBench v264%
MRCRv266%
gpqaDiamond78.7%
hle11.2%
KnowledgeSarvam 105B wins
MMLU64%90.6%
GPQA63%
SuperGPQA61%
MMLU-Pro65%81.7%
HLE7%
FrontierScience57%
SimpleQA61%
Instruction FollowingSarvam 105B wins
IFEval79%84.8%
Multilingual
MGSM75%
MMLU-ProX73%
MathematicsSarvam 105B wins
AIME 202364%
AIME 202466%
AIME 202565%88.3%
HMMT Feb 202360%
HMMT Feb 202462%
HMMT Feb 202561%
BRUMO 202563%
MATH-50073%98.6%
HMMT Feb 202585.8%
HMMT Nov 202585.8%
Frequently Asked Questions (6)

Which is better, Grok Code Fast 1 or Sarvam 105B?

Sarvam 105B is ahead overall, 60 to 56. The biggest single separator in this matchup is MMLU, where the scores are 64% and 90.6%.

Which is better for knowledge tasks, Grok Code Fast 1 or Sarvam 105B?

Sarvam 105B has the edge for knowledge tasks in this comparison, averaging 81.7 versus 49. Inside this category, MMLU is the benchmark that creates the most daylight between them.

Which is better for coding, Grok Code Fast 1 or Sarvam 105B?

Grok Code Fast 1 has the edge for coding in this comparison, averaging 63.3 versus 45. Inside this category, SWE-bench Verified is the benchmark that creates the most daylight between them.

Which is better for math, Grok Code Fast 1 or Sarvam 105B?

Sarvam 105B has the edge for math in this comparison, averaging 92.3 versus 66.3. Inside this category, MATH-500 is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, Grok Code Fast 1 or Sarvam 105B?

Grok Code Fast 1 has the edge for agentic tasks in this comparison, averaging 55.7 versus 49.5. Inside this category, BrowseComp is the benchmark that creates the most daylight between them.

Which is better for instruction following, Grok Code Fast 1 or Sarvam 105B?

Sarvam 105B has the edge for instruction following in this comparison, averaging 84.8 versus 79. Inside this category, IFEval is the benchmark that creates the most daylight between them.

Last updated: April 3, 2026

Weekly LLM Benchmark Digest

Get notified when new models drop, benchmark scores change, or the leaderboard shifts. One email per week.

Free. No spam. Unsubscribe anytime. We only store derived location metadata for consent routing.