The AI arena is free today

Open Superagent

Kimi K2 0905 vs Ministral 3 (8B Instruct 2512)

Kimi K2 0905 significantly outperforms across most benchmarks.

Moonshot AI · Mistral AI · Updated for 2026

Which is better?

Kimi K2 0905 outperforms in 1 benchmarks (MATH), while Ministral 3 (8B Instruct 2512) is better at 0 benchmarks. Kimi K2 0905 significantly outperforms across most benchmarks.

Based on current benchmark, pricing, and model metadata for 2026.

Choose Kimi K2 0905

  • you want the strongest raw capability — it leads on 1 of 1 shared benchmarks

Choose Ministral 3 (8B Instruct 2512)

  • you want the most recent training data — it shipped Dec 2025
  • you need open weights you can self-host or fine-tune

At a glance

The differences that matter most.

Kimi K2 0905
Moonshot AI
Ministral 3 (8B Instruct 2512)
Mistral AI
Benchmark wins
1 of 1
0 of 1
Input price
$0.60 / M
— / M
Output price
$2.50 / M
— / M
Context window
262,144
—
Released
Sep 2025
Dec 2025
License
Proprietary
Apache 2.0

Performance Benchmarks

Comparative analysis across standard metrics

1 benchmarks

Kimi K2 0905 outperforms in 1 benchmarks (MATH), while Ministral 3 (8B Instruct 2512) is better at 0 benchmarks.

Kimi K2 0905 significantly outperforms across most benchmarks.

Tue Aug 25 2026 • llm-stats.com

Arena Performance

Playground indexes and blind preference scores

Model Size

Parameter count comparison

992.0B diff

Kimi K2 0905 has 992.0B more parameters than Ministral 3 (8B Instruct 2512), making it 12400.0% larger.

Moonshot AI
Kimi K2 0905
1.0Tparameters
Mistral AI
Ministral 3 (8B Instruct 2512)
8.0Bparameters
1000.0B
Kimi K2 0905
8.0B
Ministral 3 (8B Instruct 2512)

Context Window

Maximum input and output token capacity

Only Kimi K2 0905 specifies input context (262,144 tokens). Only Kimi K2 0905 specifies output context (262,144 tokens).

Moonshot AI
Kimi K2 0905
Input262,144 tokens
Output262,144 tokens
Mistral AI
Ministral 3 (8B Instruct 2512)
Input- tokens
Output- tokens
Tue Aug 25 2026 • llm-stats.com

Input Capabilities

Supported data types and modalities

Ministral 3 (8B Instruct 2512) supports multimodal inputs, whereas Kimi K2 0905 does not.

Ministral 3 (8B Instruct 2512) can handle both text and other forms of data like images, making it suitable for multimodal applications.

Kimi K2 0905

Text
Images
Audio
Video

Ministral 3 (8B Instruct 2512)

Text
Images
Audio
Video

License

Usage and distribution terms

Kimi K2 0905 is licensed under a proprietary license, while Ministral 3 (8B Instruct 2512) uses Apache 2.0.

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

Kimi K2 0905

Proprietary

Closed source

Ministral 3 (8B Instruct 2512)

Apache 2.0

Open weights

Release Timeline

When each model was launched

Kimi K2 0905 was released on 2025-09-05, while Ministral 3 (8B Instruct 2512) was released on 2025-12-04.

Ministral 3 (8B Instruct 2512) is 3 months newer than Kimi K2 0905.

Kimi K2 0905

Sep 5, 2025

11 months ago

Ministral 3 (8B Instruct 2512)

Dec 4, 2025

8 months ago

3mo 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 0905 and Ministral 3 (8B Instruct 2512) side-by-side, then vote on the output you prefer.

Kimi K2 0905
✓ Preferred
Ministral 3 (8B Instruct 2512)
Open in Playground

FAQ

Common questions about Kimi K2 0905 vs Ministral 3 (8B Instruct 2512).

Which is better, Kimi K2 0905 or Ministral 3 (8B Instruct 2512)?

Kimi K2 0905 significantly outperforms across most benchmarks. Kimi K2 0905 is made by Moonshot AI and Ministral 3 (8B Instruct 2512) is made by Mistral AI. The best choice depends on your use case — compare their benchmark scores, pricing, and capabilities above.

How does Kimi K2 0905 compare to Ministral 3 (8B Instruct 2512) in benchmarks?

Kimi K2 0905 scores HumanEval: 94.5%, MMLU: 90.2%, MATH: 89.1%, MMLU-Pro: 82.5%, GPQA: 75.8%. Ministral 3 (8B Instruct 2512) scores MATH: 87.6%, Wild Bench: 66.8%, Arena Hard: 50.9%, MM-MT-Bench: 8.1%.

What are the context window sizes for Kimi K2 0905 and Ministral 3 (8B Instruct 2512)?

Kimi K2 0905 supports 262K tokens and Ministral 3 (8B Instruct 2512) 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 0905 and Ministral 3 (8B Instruct 2512)?

Key differences include multimodal support (no vs yes), licensing (Proprietary vs Apache 2.0). See the full comparison above for benchmark-by-benchmark results.

Who makes Kimi K2 0905 and Ministral 3 (8B Instruct 2512)?

Kimi K2 0905 is developed by Moonshot AI and Ministral 3 (8B Instruct 2512) is developed by Mistral AI.

Related comparisons