The AI arena is free today

Open Superagent

Gemma 4 31B vs Kimi K2-Thinking-0905

Gemma 4 31B and Kimi K2-Thinking-0905 are closely matched at 33.1 and 36.0 on the LLM Stats Score. Gemma 4 31B is 5.6x cheaper per token.

Google · Moonshot AI · Updated for 2026

Which is better?

Gemma 4 31B and Kimi K2-Thinking-0905 are closely matched on the overall LLM Stats Score at 33.1 and 36.0.

In the 4 individual benchmarks reported for both models, Kimi K2-Thinking-0905 wins 3; this is a narrower head-to-head signal than the composite indexes.

On price, Gemma 4 31B is roughly 5.6x cheaper per token on a blended 3:1 input/output basis, which adds up quickly at production volume.

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

Choose Gemma 4 31B

  • cost matters — it's about 5.6x cheaper per token
  • you want the most recent training data — it shipped Apr 2026

Choose Kimi K2-Thinking-0905

  • you value its reported benchmark strengths — it wins 3 of 4 exact shared results

At a glance

The differences that matter most.

Core performance indexes
33.1
#101
36.0
#81
33.6
#94
36.2
#74
13.6
#91
16.0
#80
Cost, coverage & limits
Benchmark wins
1 of 4
3 of 4
Input price
$0.09 / M
$0.47 / M
Output price
$0.34 / M
$2.00 / M
Context window
262,144
262,144

Capability indexes

Additional strengths measured across groups of related public benchmarks

1 shared
Index
Gemma 4 31B
Kimi K2-Thinking-0905
29.4#84
38.0#27
Conservative TrueSkill rating · higher is betterHow scores work

Individual benchmarks

12 reported for Gemma 4 31B · 21 for Kimi K2-Thinking-0905

4 shared

Gemma 4 31B outperforms in 1 benchmarks (MMLU-Pro), while Kimi K2-Thinking-0905 is better at 3 benchmarks (GPQA, Humanity's Last Exam, LiveCodeBench v6).

Kimi K2-Thinking-0905 shows notably better performance in the majority of benchmarks.

Mon Sep 21 2026 • llm-stats.com

Human preference

Blind head-to-head votes and playground preference scores

Pricing Analysis

Price comparison per million tokens

Gemma 4 31B costs less

For input processing, Gemma 4 31B ($0.09/1M tokens) is 5.2x cheaper than Kimi K2-Thinking-0905 ($0.47/1M tokens).

For output processing, Gemma 4 31B ($0.34/1M tokens) is 5.9x cheaper than Kimi K2-Thinking-0905 ($2.00/1M tokens).

In conclusion, Kimi K2-Thinking-0905 is more expensive than Gemma 4 31B.*

* Using a 3:1 ratio of input to output tokens

Lowest available price from all providers
Mon Sep 21 2026 • llm-stats.com
Google
Gemma 4 31B
Input tokens$0.09
Output tokens$0.34
Best providerDeepinfra
Moonshot AI
Kimi K2-Thinking-0905
Input tokens$0.47
Output tokens$2.00
Best providerDeepinfra
Notice missing or incorrect data?Start an Issue

Model Size

Parameter count comparison

969.3B diff

Kimi K2-Thinking-0905 has 969.3B more parameters than Gemma 4 31B, making it 3157.3% larger.

Google
Gemma 4 31B
30.7Bparameters
Moonshot AI
Kimi K2-Thinking-0905
1.0Tparameters
30.7B
Gemma 4 31B
1000.0B
Kimi K2-Thinking-0905

Context Window

Maximum input and output token capacity

Both models have the same input context window of 262,144 tokens. Both models can generate responses up to 262,144 tokens.

Google
Gemma 4 31B
Input262,144 tokens
Output262,144 tokens
Moonshot AI
Kimi K2-Thinking-0905
Input262,144 tokens
Output262,144 tokens
Mon Sep 21 2026 • llm-stats.com

Input capabilities

Documented input modalities across available providers

Gemma 4 31B supports multimodal inputs, whereas Kimi K2-Thinking-0905 does not.

Gemma 4 31B can handle both text and other forms of data like images, making it suitable for multimodal applications.

Gemma 4 31B

Text
Images
Audio
Video

Kimi K2-Thinking-0905

Text
Images
Audio
Video

License

Usage and distribution terms

Gemma 4 31B is licensed under Apache 2.0, while Kimi K2-Thinking-0905 uses MIT.

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

Gemma 4 31B

Apache 2.0

Open weights

Kimi K2-Thinking-0905

MIT

Open weights

Release Timeline

When each model was launched

Gemma 4 31B was released on 2026-04-02, while Kimi K2-Thinking-0905 was released on 2025-09-05.

Gemma 4 31B is 7 months newer than Kimi K2-Thinking-0905.

Gemma 4 31B

Apr 2, 2026

5 months ago

6mo newer
Kimi K2-Thinking-0905

Sep 5, 2025

1.0 years ago

Knowledge Cutoff

When training data ends

Gemma 4 31B has a documented knowledge cutoff of 2025-01-01, while Kimi K2-Thinking-0905's cutoff date is not specified.

We can confirm Gemma 4 31B's training data extends to 2025-01-01, but cannot make a direct comparison without Kimi K2-Thinking-0905's cutoff date.

Gemma 4 31B

Jan 2025

Kimi K2-Thinking-0905

Provider Availability

Gemma 4 31B is available from DeepInfra, FriendliAI, Novita, Together. Kimi K2-Thinking-0905 is available from DeepInfra, Novita, Fireworks.

Gemma 4 31B

deepinfra logo
Deepinfra
Input Price:Input: $0.09/1MOutput Price:Output: $0.34/1M
friendli logo
FriendliAI
Input Price:Input: $0.14/1MOutput Price:Output: $0.40/1M
novita logo
Novita
Input Price:Input: $0.14/1MOutput Price:Output: $0.40/1M
together logo
Together
Input Price:Input: $0.39/1MOutput Price:Output: $0.97/1M

Kimi K2-Thinking-0905

deepinfra logo
Deepinfra
Input Price:Input: $0.47/1MOutput Price:Output: $2.00/1M
novita logo
Novita
Input Price:Input: $0.48/1MOutput Price:Output: $2.00/1M
fireworks logo
Fireworks
Input Price:Input: $0.60/1MOutput Price:Output: $2.50/1M
* Prices shown are per million tokens

Outputs Comparison

Notice missing or incorrect data?Start an Issue discussion

Judge for yourself.

Run your own prompts against Gemma 4 31B and Kimi K2-Thinking-0905 side-by-side, then vote on the output you prefer.

Gemma 4 31B
✓ Preferred
Kimi K2-Thinking-0905
Open in Playground

FAQ

Common questions about Gemma 4 31B vs Kimi K2-Thinking-0905.

Which is better, Gemma 4 31B or Kimi K2-Thinking-0905?

Gemma 4 31B and Kimi K2-Thinking-0905 are closely matched on the LLM Stats Score at 33.1 and 36.0. Gemma 4 31B is made by Google and Kimi K2-Thinking-0905 is made by Moonshot AI. The best choice depends on your use case — compare their capability indexes, individual benchmarks, pricing, and limits above.

How does Gemma 4 31B compare to Kimi K2-Thinking-0905 in benchmarks?

Gemma 4 31B scores AIME 2026: 89.2%, MMMLU: 88.4%, t2-bench: 86.4%, MathVision: 85.6%, MMLU-Pro: 85.2%. Kimi K2-Thinking-0905 scores AIME 2025: 100.0%, HMMT 2025: 97.5%, MMLU-Redux: 94.4%, FRAMES: 87.0%, MMLU-Pro: 84.6%.

Is Gemma 4 31B cheaper than Kimi K2-Thinking-0905?

Gemma 4 31B is 5.2x cheaper for input tokens. Gemma 4 31B costs $0.09/M input and $0.34/M output via deepinfra. Kimi K2-Thinking-0905 costs $0.47/M input and $2.00/M output via deepinfra.

What are the context window sizes for Gemma 4 31B and Kimi K2-Thinking-0905?

Gemma 4 31B supports 262K tokens and Kimi K2-Thinking-0905 supports 262K tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.

What are the main differences between Gemma 4 31B and Kimi K2-Thinking-0905?

Key differences include LLM Stats Score (33.1 vs 36.0), input pricing ($0.09 vs $0.47/M), multimodal support (yes vs no), licensing (Apache 2.0 vs MIT). See the full comparison above for benchmark-by-benchmark results.

Who makes Gemma 4 31B and Kimi K2-Thinking-0905?

Gemma 4 31B is developed by Google and Kimi K2-Thinking-0905 is developed by Moonshot AI.