The AI arena is free today

Open Superagent

Gemini 3 Pro vs Kimi K2-Thinking-0905

Gemini 3 Pro and Kimi K2-Thinking-0905 are closely matched at 38.6 and 36.0 on the LLM Stats Score. Kimi K2-Thinking-0905 is 5.3x cheaper per token.

Google · Moonshot AI · Updated for 2026

Which is better?

Gemini 3 Pro and Kimi K2-Thinking-0905 are closely matched on the overall LLM Stats Score at 38.6 and 36.0.

The models split the 4 individual benchmarks reported for both models evenly.

On price, Kimi K2-Thinking-0905 is roughly 5.3x cheaper per token on a blended 3:1 input/output basis, which adds up quickly at production volume.

Gemini 3 Pro also accepts a larger context window (1,048,576 input tokens), making it the stronger choice for long documents and large codebases.

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

Choose Gemini 3 Pro

  • you process long inputs — it offers a 1,048,576 token context window
  • you want the most recent training data — it shipped Nov 2025

Choose Kimi K2-Thinking-0905

  • cost matters — it's about 5.3x cheaper per token
  • you need open weights you can self-host or fine-tune

At a glance

The differences that matter most.

Gemini 3 Pro
Google
Kimi K2-Thinking-0905
Moonshot AI
Core performance indexes
LLM Stats Score
38.6
#64
36.0
#81
Reasoning
38.6
#61
36.2
#74
Coding
24.5
#72
21.0
#91
Agents
15.3
#85
16.0
#80
Cost, coverage & limits
Benchmark wins
2 of 4
2 of 4
Input price
$2.00 / M
$0.47 / M
Output price
$12.00 / M
$2.00 / M
Context window
1,048,576
262,144

Capability indexes

Additional strengths measured across groups of related public benchmarks

1 shared
Index
Gemini 3 Pro
Kimi K2-Thinking-0905
Math
37.1#31
38.0#27
Conservative TrueSkill rating · higher is betterHow scores work

Individual benchmarks

21 reported for Gemini 3 Pro · 21 for Kimi K2-Thinking-0905

4 shared

Gemini 3 Pro outperforms in 2 benchmarks (GPQA, SWE-Bench Verified), while Kimi K2-Thinking-0905 is better at 1 benchmark (Humanity's Last Exam).

Gemini 3 Pro has a slight edge in benchmark performance.

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

Kimi K2-Thinking-0905 costs less

For input processing, Gemini 3 Pro ($2.00/1M tokens) is 4.3x more expensive than Kimi K2-Thinking-0905 ($0.47/1M tokens).

For output processing, Gemini 3 Pro ($12.00/1M tokens) is 6.0x more expensive than Kimi K2-Thinking-0905 ($2.00/1M tokens).

In conclusion, Gemini 3 Pro is more expensive than Kimi K2-Thinking-0905.*

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

Lowest available price from all providers
Mon Sep 21 2026 • llm-stats.com
Google
Gemini 3 Pro
Input tokens$2.00
Output tokens$12.00
Best providerGoogle
Moonshot AI
Kimi K2-Thinking-0905
Input tokens$0.47
Output tokens$2.00
Best providerDeepinfra
Notice missing or incorrect data?Start an Issue→

Context Window

Maximum input and output token capacity

Gemini 3 Pro accepts 1,048,576 input tokens compared to Kimi K2-Thinking-0905's 262,144 tokens. Kimi K2-Thinking-0905 can generate longer responses up to 262,144 tokens, while Gemini 3 Pro is limited to 65,536 tokens.

Google
Gemini 3 Pro
Input1,048,576 tokens
Output65,536 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

Gemini 3 Pro supports multimodal inputs, whereas Kimi K2-Thinking-0905 does not.

Gemini 3 Pro can handle both text and other forms of data like images, making it suitable for multimodal applications.

Gemini 3 Pro

Text
Images
Audio
Video

Kimi K2-Thinking-0905

Text
Images
Audio
Video

License

Usage and distribution terms

Gemini 3 Pro is licensed under a proprietary license, while Kimi K2-Thinking-0905 uses MIT.

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

Gemini 3 Pro

Proprietary

Closed source

Kimi K2-Thinking-0905

MIT

Open weights

Release Timeline

When each model was launched

Gemini 3 Pro was released on 2025-11-18, while Kimi K2-Thinking-0905 was released on 2025-09-05.

Gemini 3 Pro is 2 months newer than Kimi K2-Thinking-0905.

Gemini 3 Pro

Nov 18, 2025

10 months ago

2mo newer
Kimi K2-Thinking-0905

Sep 5, 2025

1.0 years ago

Knowledge Cutoff

When training data ends

Gemini 3 Pro has a documented knowledge cutoff of 2025-01-31, while Kimi K2-Thinking-0905's cutoff date is not specified.

We can confirm Gemini 3 Pro's training data extends to 2025-01-31, but cannot make a direct comparison without Kimi K2-Thinking-0905's cutoff date.

Gemini 3 Pro

Jan 2025

Kimi K2-Thinking-0905

—

Provider Availability

Gemini 3 Pro is available from Google. Kimi K2-Thinking-0905 is available from DeepInfra, Novita, Fireworks.

Gemini 3 Pro

google logo
Google
Input Price:Input: $2.00/1MOutput Price:Output: $12.00/1M

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 Gemini 3 Pro and Kimi K2-Thinking-0905 side-by-side, then vote on the output you prefer.

Gemini 3 Pro
✓ Preferred
Kimi K2-Thinking-0905
Open in Playground

FAQ

Common questions about Gemini 3 Pro vs Kimi K2-Thinking-0905.

Which is better, Gemini 3 Pro or Kimi K2-Thinking-0905?

Gemini 3 Pro and Kimi K2-Thinking-0905 are closely matched on the LLM Stats Score at 38.6 and 36.0. Gemini 3 Pro 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 Gemini 3 Pro compare to Kimi K2-Thinking-0905 in benchmarks?

Gemini 3 Pro scores AIME 2025: 100.0%, Vending-Bench 2: 100.0%, Global PIQA: 93.4%, GPQA: 91.9%, MMMLU: 91.8%. 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 Gemini 3 Pro cheaper than Kimi K2-Thinking-0905?

Kimi K2-Thinking-0905 is 4.3x cheaper for input tokens. Gemini 3 Pro costs $2.00/M input and $12.00/M output via google. Kimi K2-Thinking-0905 costs $0.47/M input and $2.00/M output via deepinfra.

What are the context window sizes for Gemini 3 Pro and Kimi K2-Thinking-0905?

Gemini 3 Pro supports 1.0M 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 Gemini 3 Pro and Kimi K2-Thinking-0905?

Key differences include LLM Stats Score (38.6 vs 36.0), context window (1.0M vs 262K), input pricing ($2.00 vs $0.47/M), multimodal support (yes vs no), licensing (Proprietary vs MIT). See the full comparison above for benchmark-by-benchmark results.

Who makes Gemini 3 Pro and Kimi K2-Thinking-0905?

Gemini 3 Pro is developed by Google and Kimi K2-Thinking-0905 is developed by Moonshot AI.

Related comparisons