The AI arena is free today

Open Superagent

Kimi K2.5 vs Sarvam-30B

Kimi K2.5 leads the LLM Stats Score 38.8 to 19.6.

Moonshot AI · Sarvam AI · Updated for 2026

Which is better?

Kimi K2.5 leads the overall LLM Stats Score 38.8 to 19.6, ranking #62 overall.

In the 7 individual benchmarks reported for both models, Kimi K2.5 wins 6; this is a narrower head-to-head signal than the composite indexes.

Based on current LLM Stats indexes, shared benchmarks, pricing, and model metadata for 2026.

Choose Kimi K2.5

  • overall performance matters — it scores 38.8 and ranks #62 on LLM Stats
  • your work emphasizes reasoning and coding — it leads those capability indexes
  • you value its reported benchmark strengths — it wins 6 of 7 exact shared results

Choose Sarvam-30B

  • you want the most recent training data — it shipped Mar 2026

At a glance

The differences that matter most.

Core performance indexes
38.8
#62
19.6
#202
38.7
#60
19.3
#200
25.2
#70
5.9
#193
18.3
#69
-1.4
#176
Cost, coverage & limits
Benchmark wins
6 of 7
1 of 7
Input price
$0.60 / M
— / M
Output price
$3.00 / M
— / M
Context window
262,100

Capability indexes

Additional strengths measured across groups of related public benchmarks

1 shared
Index
Kimi K2.5
Sarvam-30B
36.6#36
23.6#125
Conservative TrueSkill rating · higher is betterHow scores work

Individual benchmarks

40 reported for Kimi K2.5 · 14 for Sarvam-30B

7 shared

Kimi K2.5 outperforms in 6 benchmarks (BrowseComp, GPQA, HMMT 2025, LiveCodeBench v6, MMLU-Pro, SWE-Bench Verified), while Sarvam-30B is better at 1 benchmark (AIME 2025).

Kimi K2.5 significantly outperforms across most benchmarks.

Wed Sep 16 2026 • llm-stats.com

Human preference

Blind head-to-head votes and playground preference scores

Model Size

Parameter count comparison

970.0B diff

Kimi K2.5 has 970.0B more parameters than Sarvam-30B, making it 3233.3% larger.

Moonshot AI
Kimi K2.5
1.0Tparameters
Sarvam AI
Sarvam-30B
30.0Bparameters
1000.0B
Kimi K2.5
30.0B
Sarvam-30B

Context Window

Maximum input and output token capacity

Only Kimi K2.5 specifies input context (262,100 tokens). Only Kimi K2.5 specifies output context (262,100 tokens).

Moonshot AI
Kimi K2.5
Input262,100 tokens
Output262,100 tokens
Sarvam AI
Sarvam-30B
Input- tokens
Output- tokens
Wed Sep 16 2026 • llm-stats.com

Input capabilities

Documented input modalities across available providers

Kimi K2.5 supports multimodal inputs, whereas Sarvam-30B does not.

Kimi K2.5 can handle both text and other forms of data like images, making it suitable for multimodal applications.

Kimi K2.5

Text
Images
Audio
Video

Sarvam-30B

Text
Images
Audio
Video

License

Usage and distribution terms

Kimi K2.5 is licensed under MIT, while Sarvam-30B uses Apache 2.0.

License differences may affect how you can use these models in commercial or open-source projects.

Kimi K2.5

MIT

Open weights

Sarvam-30B

Apache 2.0

Open weights

Release Timeline

When each model was launched

Kimi K2.5 was released on 2026-01-27, while Sarvam-30B was released on 2026-03-06.

Sarvam-30B is 1 month newer than Kimi K2.5.

Kimi K2.5

Jan 27, 2026

7 months ago

Sarvam-30B

Mar 6, 2026

6 months ago

1mo newer

Knowledge Cutoff

When training data ends

Neither model specifies a knowledge cutoff date.

Unable to compare the recency of their training data.

No cutoff dates available

Outputs Comparison

Notice missing or incorrect data?Start an Issue discussion

Judge for yourself.

Run your own prompts against Kimi K2.5 and Sarvam-30B side-by-side, then vote on the output you prefer.

Kimi K2.5
✓ Preferred
Sarvam-30B
Open in Playground

FAQ

Common questions about Kimi K2.5 vs Sarvam-30B.

Which is better, Kimi K2.5 or Sarvam-30B?

Kimi K2.5 leads the LLM Stats Score 38.8 to 19.6. Kimi K2.5 is made by Moonshot AI and Sarvam-30B is made by Sarvam AI. The best choice depends on your use case — compare their capability indexes, individual benchmarks, pricing, and limits above.

How does Kimi K2.5 compare to Sarvam-30B in benchmarks?

Kimi K2.5 scores AIME 2025: 96.1%, HMMT 2025: 95.4%, InfoVQAtest: 92.6%, OCRBench: 92.3%, MathVista-Mini: 90.1%. Sarvam-30B scores MATH-500: 97.0%, AIME 2025: 96.7%, MBPP: 92.7%, HumanEval: 92.1%, MMLU: 85.1%.

What are the context window sizes for Kimi K2.5 and Sarvam-30B?

Kimi K2.5 supports 262K tokens and Sarvam-30B supports an unknown number of tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.

What are the main differences between Kimi K2.5 and Sarvam-30B?

Key differences include LLM Stats Score (38.8 vs 19.6), multimodal support (yes vs no), licensing (MIT vs Apache 2.0). See the full comparison above for benchmark-by-benchmark results.

Who makes Kimi K2.5 and Sarvam-30B?

Kimi K2.5 is developed by Moonshot AI and Sarvam-30B is developed by Sarvam AI.