The AI arena is free today

Open Superagent

DeepSeek-V3 0324 vs Kimi K2-Instruct-0905

DeepSeek-V3 0324 and Kimi K2-Instruct-0905 are closely matched at 13.4 and 21.6 on the LLM Stats Score.

DeepSeek · Moonshot AI · Updated for 2026

Which is better?

DeepSeek-V3 0324 and Kimi K2-Instruct-0905 are closely matched on the overall LLM Stats Score at 13.4 and 21.6.

In the 5 individual benchmarks reported for both models, Kimi K2-Instruct-0905 wins 4; 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 0324

  • you want predictable pricing at $0.24/M input and $0.90/M output

Choose Kimi K2-Instruct-0905

  • you value its reported benchmark strengths — it wins 4 of 5 exact shared results
  • you want the most recent training data — it shipped Sep 2025

At a glance

The differences that matter most.

DeepSeek-V3 0324
DeepSeek
Kimi K2-Instruct-0905
Moonshot AI
Core performance indexes
LLM Stats Score
13.4
#251
21.6
#192
Reasoning
13.6
#241
21.8
#182
Coding
4.0
#219
9.7
#175
Cost, coverage & limits
Benchmark wins
1 of 5
4 of 5
Input price
$0.24 / M
— / M
Output price
$0.90 / M
— / M
Context window
163,840
—

Capability indexes

Additional strengths measured across groups of related public benchmarks

1 shared
Index
DeepSeek-V3 0324
Kimi K2-Instruct-0905
Math
15.8#213
21.5#145
Conservative TrueSkill rating · higher is betterHow scores work

Individual benchmarks

5 reported for DeepSeek-V3 0324 · 29 for Kimi K2-Instruct-0905

5 shared

DeepSeek-V3 0324 outperforms in 1 benchmarks (MMLU-Pro), while Kimi K2-Instruct-0905 is better at 4 benchmarks (AIME 2024, GPQA, LiveCodeBench, MATH-500).

Kimi K2-Instruct-0905 significantly outperforms across most benchmarks.

Wed Sep 23 2026 • llm-stats.com

Human preference

Blind head-to-head votes and playground preference scores

Model Size

Parameter count comparison

329.0B diff

Kimi K2-Instruct-0905 has 329.0B more parameters than DeepSeek-V3 0324, making it 49.0% larger.

DeepSeek
DeepSeek-V3 0324
671.0Bparameters
Moonshot AI
Kimi K2-Instruct-0905
1.0Tparameters
671.0B
DeepSeek-V3 0324
1000.0B
Kimi K2-Instruct-0905

Context Window

Maximum input and output token capacity

Only DeepSeek-V3 0324 specifies input context (163,840 tokens). Only DeepSeek-V3 0324 specifies output context (163,840 tokens).

DeepSeek
DeepSeek-V3 0324
Input163,840 tokens
Output163,840 tokens
Moonshot AI
Kimi K2-Instruct-0905
Input- tokens
Output- tokens
Wed Sep 23 2026 • llm-stats.com

License

Usage and distribution terms

DeepSeek-V3 0324 is licensed under MIT + Model License (Commercial use allowed), while Kimi K2-Instruct-0905 uses MIT.

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

DeepSeek-V3 0324

MIT + Model License (Commercial use allowed)

Open weights

Kimi K2-Instruct-0905

MIT

Open weights

Release Timeline

When each model was launched

DeepSeek-V3 0324 was released on 2025-03-25, while Kimi K2-Instruct-0905 was released on 2025-09-05.

Kimi K2-Instruct-0905 is 5 months newer than DeepSeek-V3 0324.

DeepSeek-V3 0324

Mar 25, 2025

1.5 years ago

Kimi K2-Instruct-0905

Sep 5, 2025

1.0 years ago

5mo 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 0324 and Kimi K2-Instruct-0905 side-by-side, then vote on the output you prefer.

DeepSeek-V3 0324
✓ Preferred
Kimi K2-Instruct-0905
Open in Playground

FAQ

Common questions about DeepSeek-V3 0324 vs Kimi K2-Instruct-0905.

Which is better, DeepSeek-V3 0324 or Kimi K2-Instruct-0905?

DeepSeek-V3 0324 and Kimi K2-Instruct-0905 are closely matched on the LLM Stats Score at 13.4 and 21.6. DeepSeek-V3 0324 is made by DeepSeek and Kimi K2-Instruct-0905 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 0324 compare to Kimi K2-Instruct-0905 in benchmarks?

DeepSeek-V3 0324 scores MATH-500: 94.0%, MMLU-Pro: 81.2%, GPQA: 68.4%, AIME 2024: 59.4%, LiveCodeBench: 49.2%. Kimi K2-Instruct-0905 scores MATH-500: 97.4%, MMLU-Redux: 92.7%, IFEval: 89.8%, AutoLogi: 89.5%, MMLU: 89.5%.

What are the context window sizes for DeepSeek-V3 0324 and Kimi K2-Instruct-0905?

DeepSeek-V3 0324 supports 164K tokens and Kimi K2-Instruct-0905 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 0324 and Kimi K2-Instruct-0905?

Key differences include LLM Stats Score (13.4 vs 21.6), licensing (MIT + Model License (Commercial use allowed) vs MIT). See the full comparison above for benchmark-by-benchmark results.

Who makes DeepSeek-V3 0324 and Kimi K2-Instruct-0905?

DeepSeek-V3 0324 is developed by DeepSeek and Kimi K2-Instruct-0905 is developed by Moonshot AI.

Related comparisons