DeepSeek V4 Flash 0731 vs Kimi K2 0905

Price, context window & benchmark comparison

DeepSeek

DeepSeek V4 Flash 0731

DeepSeek
$0.0434 / $0.30
Input / Output · per 1M tokens
VS
Moonshot AI

Kimi K2 0905

Moonshot AI
$0.60 / $2.50
Input / Output · per 1M tokens
Compare different models
DeepSeek
Moonshot AI
Verdict: DeepSeek V4 Flash 0731 is 10.0× cheaper than Kimi K2 0905 on a typical 3:1 input/output mix. DeepSeek V4 Flash 0731 has the larger context window (1M vs 262K).
DeepSeek V4 Flash 0731Kimi K2 0905
Input price / 1M$0.0434$0.60
Output price / 1M$0.30$2.50
Blended price (3:1)$0.11$1.07
Cached input / 1M$0.035—
Context window1M262K
Max output944K98K
Intelligence index——
Coding index69.1—
Providers27—
Best provider uptime (30m)100.0%—
Vision inputNoNo
ReasoningYesNo
Tool callingYesYes
Released2026-07-312025-09-04
Retirement——

Cost calculator

FAQ

Which is cheaper, DeepSeek V4 Flash 0731 or Kimi K2 0905?

DeepSeek V4 Flash 0731 is 10.0× cheaper than Kimi K2 0905 on a typical 3:1 input/output mix.

DeepSeek V4 Flash 0731 vs Kimi K2 0905: which has a bigger context window?

DeepSeek V4 Flash 0731: 1,048,576 tokens. Kimi K2 0905: 262,144 tokens.

How much would 1 million requests cost?

At 2,000 input + 500 output tokens per request: DeepSeek V4 Flash 0731 ≈ $236.80, Kimi K2 0905 ≈ $2,450.

More comparisons