The AI arena is free today

Open Superagent

Codestral-22B vs Kimi K3.1 Pro

Kimi K3.1 Pro leads the LLM Stats Score 55.5 to -0.4.

Mistral AI · Moonshot AI · Updated for 2026

Which is better?

Kimi K3.1 Pro leads the overall LLM Stats Score 55.5 to -0.4, ranking #5 overall.

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

Choose Codestral-22B

  • you need open weights you can self-host or fine-tune

Choose Kimi K3.1 Pro

  • overall performance matters — it scores 55.5 and ranks #5 on LLM Stats
  • your work emphasizes reasoning and coding — it leads those capability indexes
  • you want the most recent training data — it shipped Oct 2026

At a glance

The differences that matter most.

Codestral-22B
Mistral AI
Kimi K3.1 Pro
Moonshot AI
Core performance indexes
LLM Stats Score
-0.4
#363
55.5
#5
Reasoning
-0.5
#356
54.3
#6
Coding
2.4
#259
42.7
#9
Cost, coverage & limits
Benchmark wins
—
—
Input price
— / M
$2.00 / M
Output price
— / M
$10.00 / M
Context window
—
1,048,576

Individual benchmarks

7 reported for Codestral-22B · 30 for Kimi K3.1 Pro

No common benchmarks found

Codestral-22B and Kimi K3.1 Prodon't have any common benchmark datasets to compare. They may have been evaluated on different testing suites.

Human preference

Blind head-to-head votes and playground preference scores

Context Window

Maximum input and output token capacity

Only Kimi K3.1 Pro specifies input context (1,048,576 tokens). Only Kimi K3.1 Pro specifies output context (1,048,576 tokens).

Mistral AI
Codestral-22B
Input- tokens
Output- tokens
Moonshot AI
Kimi K3.1 Pro
Input1,048,576 tokens
Output1,048,576 tokens
Fri Oct 30 2026 • llm-stats.com

Input capabilities

Documented input modalities across available providers

Kimi K3.1 Pro supports multimodal inputs, whereas Codestral-22B does not.

Kimi K3.1 Pro can handle both text and other forms of data like images, making it suitable for multimodal applications.

Codestral-22B

Text
Images
Audio
Video

Kimi K3.1 Pro

Text
Images
Audio
Video

Release Timeline

When each model was launched

Codestral-22B was released on 2024-05-29, while Kimi K3.1 Pro was released on 2026-10-23.

Kimi K3.1 Pro is 29 months newer than Codestral-22B.

Codestral-22B

May 29, 2024

2.4 years ago

Kimi K3.1 Pro

Oct 23, 2026

3 days ago

2.4yr 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?

Judge for yourself.

Run your own prompts against Codestral-22B and Kimi K3.1 Pro side-by-side, then vote on the output you prefer.

Codestral-22B
✓ Preferred
Kimi K3.1 Pro
Open in Playground

FAQ

Common questions about Codestral-22B vs Kimi K3.1 Pro.

Which is better, Codestral-22B or Kimi K3.1 Pro?

Kimi K3.1 Pro leads the LLM Stats Score 55.5 to -0.4. Codestral-22B is made by Mistral AI and Kimi K3.1 Pro 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 Codestral-22B compare to Kimi K3.1 Pro in benchmarks?

Codestral-22B scores HumanEvalFIM-Average: 91.6%, HumanEval: 81.1%, MBPP: 78.2%, Spider: 63.5%, HumanEval-Average: 61.5%. Kimi K3.1 Pro scores BabyVision: 95.4%, GPQA: 94.2%, BenchCAD (with Python tool): 92.7%, BrowseComp: 92.1%, Chartography: 90.2%.

What are the context window sizes for Codestral-22B and Kimi K3.1 Pro?

Codestral-22B supports an unknown number of tokens and Kimi K3.1 Pro supports 1.0M tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.

What are the main differences between Codestral-22B and Kimi K3.1 Pro?

Key differences include LLM Stats Score (-0.4 vs 55.5), multimodal support (no vs yes), licensing (MNPL-0.1 vs Unknown). See the full comparison above for benchmark-by-benchmark results.

Who makes Codestral-22B and Kimi K3.1 Pro?

Codestral-22B is developed by Mistral AI and Kimi K3.1 Pro is developed by Moonshot AI.

Related comparisons