Kimi K2.5 vs Mistral Small 4 Comparison

Comparing Kimi K2.5 and Mistral Small 4 across benchmarks, pricing, and capabilities.

Performance Benchmarks

Comparative analysis across standard metrics

5 benchmarks

Kimi K2.5 outperforms in 4 benchmarks (AIME 2025, GPQA, MMLU-Pro, MMMU-Pro), while Mistral Small 4 is better at 1 benchmark (AA-LCR).

Kimi K2.5 significantly outperforms across most benchmarks.

Wed Mar 18 2026 • llm-stats.com

Arena Performance

Human preference votes

Pricing Analysis

Price comparison per million tokens

Mistral Small 4 costs less

For input processing, Kimi K2.5 ($0.60/1M tokens) is 4.0x more expensive than Mistral Small 4 ($0.15/1M tokens).

For output processing, Kimi K2.5 ($2.50/1M tokens) is 4.2x more expensive than Mistral Small 4 ($0.60/1M tokens).

In conclusion, Kimi K2.5 is more expensive than Mistral Small 4.*

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

Lowest available price from all providers
Wed Mar 18 2026 • llm-stats.com
Moonshot AI
Kimi K2.5
Input tokens$0.60
Output tokens$2.50
Best providerFireworks
Mistral AI
Mistral Small 4
Input tokens$0.15
Output tokens$0.60
Best providerMistral
Notice missing or incorrect data?Start an Issue

Model Size

Parameter count comparison

881.0B diff

Kimi K2.5 has 881.0B more parameters than Mistral Small 4, making it 740.3% larger.

Moonshot AI
Kimi K2.5
1000.0Bparameters
Mistral AI
Mistral Small 4
119.0Bparameters
1000.0B
Kimi K2.5
119.0B
Mistral Small 4

Context Window

Maximum input and output token capacity

Kimi K2.5 accepts 262,100 input tokens compared to Mistral Small 4's 256,000 tokens. Kimi K2.5 can generate longer responses up to 262,100 tokens, while Mistral Small 4 is limited to 256,000 tokens.

Moonshot AI
Kimi K2.5
Input262,100 tokens
Output262,100 tokens
Mistral AI
Mistral Small 4
Input256,000 tokens
Output256,000 tokens
Wed Mar 18 2026 • llm-stats.com

Input Capabilities

Supported data types and modalities

Both Kimi K2.5 and Mistral Small 4 support multimodal inputs.

They are both capable of processing various types of data, offering versatility in application.

Kimi K2.5

Text
Images
Audio
Video

Mistral Small 4

Text
Images
Audio
Video

License

Usage and distribution terms

Kimi K2.5 is licensed under MIT, while Mistral Small 4 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

Mistral Small 4

Apache 2.0

Open weights

Release Timeline

When each model was launched

Kimi K2.5 was released on 2026-01-27, while Mistral Small 4 was released on 2026-03-16.

Mistral Small 4 is 2 months newer than Kimi K2.5.

Kimi K2.5

Jan 27, 2026

1 months ago

Mistral Small 4

Mar 16, 2026

2 days 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

Provider Availability

Kimi K2.5 is available from Fireworks. Mistral Small 4 is available from Mistral AI. The availability of providers can affect quality of the model and reliability.

Kimi K2.5

fireworks logo
Fireworks
Input Price:Input: $0.60/1MOutput Price:Output: $2.50/1M

Mistral Small 4

mistral logo
Mistral
Input Price:Input: $0.15/1MOutput Price:Output: $0.60/1M
* Prices shown are per million tokens

Outputs Comparison

Notice missing or incorrect data?Start an Issue discussion

Key Takeaways

Larger context window (262,100 tokens)
Higher AIME 2025 score (96.1% vs 83.8%)
Higher GPQA score (87.6% vs 71.2%)
Higher MMLU-Pro score (87.1% vs 78.0%)
Higher MMMU-Pro score (78.5% vs 60.0%)
Less expensive input tokens
Less expensive output tokens
Higher AA-LCR score (71.2% vs 70.0%)

Detailed Comparison

AI Model Comparison Table
Feature
Moonshot AI
Kimi K2.5
Mistral AI
Mistral Small 4