LV
LLMversus

Codestral 22B vs DeepSeek V2.5: Pricing, Benchmarks & Verdict (2026)

Pricing verified Apr 8, 2026

⚡ Quick Answer

Compare Codestral 22B and DeepSeek V2.5 across pricing, benchmarks, and capabilities.

Updated: April 8, 2026 · ✓ Pricing verified

Side-by-Side Comparison

FeatureCodestral 22BDeepSeek V2.5
ProviderMistral AIDeepSeek
Input Price / 1M tokens$0.200$0.140
Output Price / 1M tokens$0.600$0.280
Context Window
256K
128K
Max Output Tokens
4,096
4,096
Arena ELON/A1,200
Coding ELO
1,250
1,220
TTFT (ms)
150
150
Tokens/sec
100
100
MultimodalNoNo
JSON ModeYesYes
Function CallingYesYes
VisionNoNo
When to Use Codestral 22B

Codestral 22B excels at coding, long-context tasks.

Strengths:

  • Best-in-class code generation
  • Huge context for repos
  • Function calling strength

Best for:

codinglong-context
When to Use DeepSeek V2.5

DeepSeek V2.5 excels at coding, general-purpose tasks.

Strengths:

  • Excellent value
  • Strong code generation
  • Good quality

Best for:

codinggeneral-purpose

Related Comparisons

Claude Opus 4 vs DeepSeek V2.5Claude Opus 4 vs Codestral 22BDeepSeek V2.5 vs Gemini 2 5 ProCodestral 22B vs Gemini 2 5 ProDeepSeek V2.5 vs O3Codestral 22B vs O3Deepseek R1 vs DeepSeek V2.5Codestral 22B vs Deepseek R1DeepSeek V2.5 vs O1Codestral 22B vs O1DeepSeek V2.5 vs Qwen 3 235bCodestral 22B vs Qwen 3 235bDeepSeek V2.5 vs Gemini Exp 1206Codestral 22B vs Gemini Exp 1206DeepSeek V2.5 vs Gpt 4 5Codestral 22B vs Gpt 4 5DeepSeek V2.5 vs Groq Deepseek R1Codestral 22B vs Groq Deepseek R1DeepSeek V2.5 vs Llama 4 MaverickCodestral 22B vs Llama 4 Maverick