The AI arena is free today

Open Superagent

DeepSeek-V4-Flash-0731 vs Kimi K2-Instruct-0905

Comparing DeepSeek-V4-Flash-0731 and Kimi K2-Instruct-0905 across benchmarks, pricing, and capabilities.

DeepSeek · Moonshot AI · Updated for 2026

Which is better?

DeepSeek-V4-Flash-0731 and Kimi K2-Instruct-0905 trade strengths across price, capabilities, and technical limits. The better choice depends on the workload.

Based on current benchmark, pricing, and model metadata for 2026.

Choose DeepSeek-V4-Flash-0731

  • you want the most recent training data — it shipped Jul 2026

Choose Kimi K2-Instruct-0905

  • you are already invested in the Moonshot AI ecosystem

At a glance

The differences that matter most.

DeepSeek-V4-Flash-0731
DeepSeek
Kimi K2-Instruct-0905
Moonshot AI
Benchmark wins
—
—
Input price
$0.09 / M
— / M
Output price
$0.18 / M
— / M
Context window
1,048,576
—
Released
Jul 2026
Sep 2025
License
MIT
MIT

Performance Benchmarks

Comparative analysis across standard metrics

No common benchmarks found

DeepSeek-V4-Flash-0731 and Kimi K2-Instruct-0905don't have any common benchmark datasets to compare. They may have been evaluated on different testing suites.

Arena Performance

Playground indexes and blind preference scores

Model Size

Parameter count comparison

696.0B diff

Kimi K2-Instruct-0905 has 696.0B more parameters than DeepSeek-V4-Flash-0731, making it 228.9% larger.

DeepSeek
DeepSeek-V4-Flash-0731
304.0Bparameters
Moonshot AI
Kimi K2-Instruct-0905
1.0Tparameters
304.0B
DeepSeek-V4-Flash-0731
1000.0B
Kimi K2-Instruct-0905

Context Window

Maximum input and output token capacity

Only DeepSeek-V4-Flash-0731 specifies input context (1,048,576 tokens). Only DeepSeek-V4-Flash-0731 specifies output context (65,536 tokens).

DeepSeek
DeepSeek-V4-Flash-0731
Input1,048,576 tokens
Output65,536 tokens
Moonshot AI
Kimi K2-Instruct-0905
Input- tokens
Output- tokens
Mon Aug 24 2026 • llm-stats.com

License

Usage and distribution terms

Both models are licensed under MIT.

Both models share the same licensing terms, providing consistent usage rights.

DeepSeek-V4-Flash-0731

MIT

Open weights

Kimi K2-Instruct-0905

MIT

Open weights

Release Timeline

When each model was launched

DeepSeek-V4-Flash-0731 was released on 2026-07-31, while Kimi K2-Instruct-0905 was released on 2025-09-05.

DeepSeek-V4-Flash-0731 is 11 months newer than Kimi K2-Instruct-0905.

DeepSeek-V4-Flash-0731

Jul 31, 2026

3 weeks ago

10mo newer
Kimi K2-Instruct-0905

Sep 5, 2025

11 months ago

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-V4-Flash-0731 and Kimi K2-Instruct-0905 side-by-side, then vote on the output you prefer.

DeepSeek-V4-Flash-0731
✓ Preferred
Kimi K2-Instruct-0905
Open in Playground

FAQ

Common questions about DeepSeek-V4-Flash-0731 vs Kimi K2-Instruct-0905.

Which is better, DeepSeek-V4-Flash-0731 or Kimi K2-Instruct-0905?

DeepSeek-V4-Flash-0731 (DeepSeek) and Kimi K2-Instruct-0905 (Moonshot AI) each have strengths in different areas. Compare their benchmark scores, pricing, context windows, and capabilities above to determine which fits your needs.

How does DeepSeek-V4-Flash-0731 compare to Kimi K2-Instruct-0905 in benchmarks?

DeepSeek-V4-Flash-0731 scores Terminal-Bench 2.1: 82.7%, CyberGym: 76.7%, Toolathlon: 70.3%, DSBench-FullStack: 68.7%, DSBench-Hard: 59.6%. 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-V4-Flash-0731 and Kimi K2-Instruct-0905?

DeepSeek-V4-Flash-0731 supports 1.0M 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.

Who makes DeepSeek-V4-Flash-0731 and Kimi K2-Instruct-0905?

DeepSeek-V4-Flash-0731 is developed by DeepSeek and Kimi K2-Instruct-0905 is developed by Moonshot AI.

Related comparisons