The AI arena is free today

Open Superagent

DeepSeek-V3 vs Kimi-k1.5

DeepSeek-V3 and Kimi-k1.5 are closely matched at 15.7 and 17.2 on the LLM Stats Score.

DeepSeek · Moonshot AI · Updated for 2026

Which is better?

DeepSeek-V3 and Kimi-k1.5 are closely matched on the overall LLM Stats Score at 15.7 and 17.2.

In the 6 individual benchmarks reported for both models, Kimi-k1.5 wins 5; this is a narrower head-to-head signal than the composite indexes.

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

Choose DeepSeek-V3

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

Choose Kimi-k1.5

  • you value its reported benchmark strengths — it wins 5 of 6 exact shared results
  • you want the most recent training data — it shipped Jan 2025

At a glance

The differences that matter most.

Core performance indexes
15.7
#234
17.2
#224
14.8
#233
16.4
#221
Cost, coverage & limits
Benchmark wins
1 of 6
5 of 6
Input price
$0.27 / M
— / M
Output price
$0.89 / M
— / M
Context window
131,072
—

Capability indexes

Additional strengths measured across groups of related public benchmarks

1 shared
Index
DeepSeek-V3
Kimi-k1.5
18.0#189
20.2#161
Conservative TrueSkill rating · higher is betterHow scores work

Individual benchmarks

20 reported for DeepSeek-V3 · 9 for Kimi-k1.5

6 shared

DeepSeek-V3 outperforms in 1 benchmarks (MMLU), while Kimi-k1.5 is better at 5 benchmarks (AIME 2024, C-Eval, CLUEWSC, IFEval, MATH-500).

Kimi-k1.5 significantly outperforms across most benchmarks.

Sun Sep 27 2026 • llm-stats.com

Human preference

Blind head-to-head votes and playground preference scores

Context Window

Maximum input and output token capacity

Only DeepSeek-V3 specifies input context (131,072 tokens). Only DeepSeek-V3 specifies output context (131,072 tokens).

DeepSeek
DeepSeek-V3
Input131,072 tokens
Output131,072 tokens
Moonshot AI
Kimi-k1.5
Input- tokens
Output- tokens
Sun Sep 27 2026 • llm-stats.com

Input capabilities

Documented input modalities across available providers

Kimi-k1.5 supports multimodal inputs, whereas DeepSeek-V3 does not.

Kimi-k1.5 can handle both text and other forms of data like images, making it suitable for multimodal applications.

DeepSeek-V3

Text
Images
Audio
Video

Kimi-k1.5

Text
Images
Audio
Video

License

Usage and distribution terms

DeepSeek-V3 is licensed under MIT + Model License (Commercial use allowed), while Kimi-k1.5 uses a proprietary license.

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

DeepSeek-V3

MIT + Model License (Commercial use allowed)

Open weights

Kimi-k1.5

Proprietary

Closed source

Release Timeline

When each model was launched

DeepSeek-V3 was released on 2024-12-25, while Kimi-k1.5 was released on 2025-01-20.

Kimi-k1.5 is 1 month newer than DeepSeek-V3.

DeepSeek-V3

Dec 25, 2024

1.8 years ago

Kimi-k1.5

Jan 20, 2025

1.7 years ago

3w 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 DeepSeek-V3 and Kimi-k1.5 side-by-side, then vote on the output you prefer.

DeepSeek-V3
✓ Preferred
Kimi-k1.5
Open in Playground

FAQ

Common questions about DeepSeek-V3 vs Kimi-k1.5.

Which is better, DeepSeek-V3 or Kimi-k1.5?

DeepSeek-V3 and Kimi-k1.5 are closely matched on the LLM Stats Score at 15.7 and 17.2. DeepSeek-V3 is made by DeepSeek and Kimi-k1.5 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 DeepSeek-V3 compare to Kimi-k1.5 in benchmarks?

DeepSeek-V3 scores DROP: 91.6%, CLUEWSC: 90.9%, MATH-500: 90.2%, MMLU-Redux: 89.1%, MMLU: 88.5%. Kimi-k1.5 scores MATH-500: 96.2%, CLUEWSC: 91.4%, C-Eval: 88.3%, MMLU: 87.4%, IFEval: 87.2%.

What are the context window sizes for DeepSeek-V3 and Kimi-k1.5?

DeepSeek-V3 supports 131K tokens and Kimi-k1.5 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 DeepSeek-V3 and Kimi-k1.5?

Key differences include LLM Stats Score (15.7 vs 17.2), multimodal support (no vs yes), licensing (MIT + Model License (Commercial use allowed) vs Proprietary). See the full comparison above for benchmark-by-benchmark results.

Who makes DeepSeek-V3 and Kimi-k1.5?

DeepSeek-V3 is developed by DeepSeek and Kimi-k1.5 is developed by Moonshot AI.