Model Comparison

Kimi K2.5 vs Qwen3.6-35B-A3B

Kimi K2.5 significantly outperforms across most benchmarks.

Want to compare interactively?Try the playground

Performance Benchmarks

Comparative analysis across standard metrics

18 benchmarks

Kimi K2.5 outperforms in 15 benchmarks (GPQA, HMMT 2025, Humanity's Last Exam, IMO-AnswerBench, LiveCodeBench v6, LVBench, MathVista-Mini, MMLU-Pro, MMMU-Pro, SimpleVQA, SWE-bench Multilingual, SWE-Bench Pro, SWE-Bench Verified, VideoMMMU, WideSearch), while Qwen3.6-35B-A3B is better at 3 benchmarks (CharXiv-R, OmniDocBench 1.5, Terminal-Bench 2.0).

Kimi K2.5 significantly outperforms across most benchmarks.

Sat Apr 18 2026 • llm-stats.com

Arena Performance

Human preference votes

Pricing Analysis

Price comparison per million tokens

Cost data unavailable.

Lowest available price from all providers
Sat Apr 18 2026 • llm-stats.com
Moonshot AI
Kimi K2.5
Input tokens$0.60
Output tokens$3.00
Best providerFireworks
Alibaba Cloud / Qwen Team
Qwen3.6-35B-A3B
Input tokens$0.00
Output tokens$0.00
Best providerUnknown Organization
Notice missing or incorrect data?Start an Issue→

Model Size

Parameter count comparison

965.0B diff

Kimi K2.5 has 965.0B more parameters than Qwen3.6-35B-A3B, making it 2757.1% larger.

Moonshot AI
Kimi K2.5
1000.0Bparameters
Alibaba Cloud / Qwen Team
Qwen3.6-35B-A3B
35.0Bparameters
1000.0B
Kimi K2.5
35.0B
Qwen3.6-35B-A3B

Context Window

Maximum input and output token capacity

Only Kimi K2.5 specifies input context (262,100 tokens). Only Kimi K2.5 specifies output context (262,100 tokens).

Moonshot AI
Kimi K2.5
Input262,100 tokens
Output262,100 tokens
Alibaba Cloud / Qwen Team
Qwen3.6-35B-A3B
Input- tokens
Output- tokens
Sat Apr 18 2026 • llm-stats.com

Input Capabilities

Supported data types and modalities

Both Kimi K2.5 and Qwen3.6-35B-A3B support multimodal inputs.

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

Kimi K2.5

Text
Images
Audio
Video

Qwen3.6-35B-A3B

Text
Images
Audio
Video

License

Usage and distribution terms

Kimi K2.5 is licensed under MIT, while Qwen3.6-35B-A3B 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

Qwen3.6-35B-A3B

Apache 2.0

Open weights

Release Timeline

When each model was launched

Kimi K2.5 was released on 2026-01-27, while Qwen3.6-35B-A3B was released on 2026-04-16.

Qwen3.6-35B-A3B is 3 months newer than Kimi K2.5.

Kimi K2.5

Jan 27, 2026

2 months ago

Qwen3.6-35B-A3B

Apr 16, 2026

2 days ago

2mo 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→

Key Takeaways

Moonshot AI
Larger context window (262,100 tokens)
Higher GPQA score (87.6% vs 86.0%)
Higher HMMT 2025 score (95.4% vs 90.7%)
Higher Humanity's Last Exam score (50.2% vs 21.4%)
Higher IMO-AnswerBench score (81.8% vs 78.9%)
Higher LiveCodeBench v6 score (85.0% vs 80.4%)
Higher LVBench score (75.9% vs 71.4%)
Higher MathVista-Mini score (90.1% vs 86.4%)
Higher MMLU-Pro score (87.1% vs 85.2%)
Higher MMMU-Pro score (78.5% vs 75.3%)
Higher SimpleVQA score (71.2% vs 58.9%)
Higher SWE-bench Multilingual score (73.0% vs 67.2%)
Higher SWE-Bench Pro score (50.7% vs 49.5%)
Higher SWE-Bench Verified score (76.8% vs 73.4%)
Higher VideoMMMU score (86.6% vs 83.7%)
Higher WideSearch score (79.0% vs 60.1%)
Alibaba Cloud / Qwen Team

Qwen3.6-35B-A3B

View details

Alibaba Cloud / Qwen Team

Higher CharXiv-R score (78.0% vs 77.5%)
Higher OmniDocBench 1.5 score (89.9% vs 88.8%)
Higher Terminal-Bench 2.0 score (51.5% vs 50.8%)

Detailed Comparison

AI Model Comparison Table
Feature
Moonshot AI
Kimi K2.5
Alibaba Cloud / Qwen Team
Qwen3.6-35B-A3B

FAQ

Common questions about Kimi K2.5 vs Qwen3.6-35B-A3B

Kimi K2.5 significantly outperforms across most benchmarks. Kimi K2.5 is made by Moonshot AI and Qwen3.6-35B-A3B is made by Alibaba Cloud / Qwen Team. The best choice depends on your use case — compare their benchmark scores, pricing, and capabilities above.
Kimi K2.5 scores AIME 2025: 96.1%, HMMT 2025: 95.4%, InfoVQAtest: 92.6%, OCRBench: 92.3%, MathVista-Mini: 90.1%. Qwen3.6-35B-A3B scores MMLU-Redux: 93.3%, MMBench-V1.1: 92.8%, AI2D: 92.7%, AIME 2026: 92.7%, RefCOCO-avg: 92.0%.
Kimi K2.5 supports 262K tokens and Qwen3.6-35B-A3B supports an unknown number of tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.
Key differences include licensing (MIT vs Apache 2.0). See the full comparison above for benchmark-by-benchmark results.
Kimi K2.5 is developed by Moonshot AI and Qwen3.6-35B-A3B is developed by Alibaba Cloud / Qwen Team.