Claude Sonnet 4.5 vs Codestral
Side-by-side benchmark comparison across coding, math, reasoning, speed, and pricing.
Claude Sonnet 4.5 by Anthropic wins on 10 of 15 benchmarks against Codestral by Mistral, which leads on 5. This head-to-head comparison covers coding, math, reasoning, speed, and pricing metrics from our benchmark data.
Category-by-Category Breakdown
General Intelligence: In general intelligence, Claude Sonnet 4.5 scores 1380 on Chatbot Arena ELO compared to Codestral's 1260, while Claude Sonnet 4.5 scores 83.0% on MMLU-Pro compared to Codestral's 68.0%.
Coding: In coding, Claude Sonnet 4.5 scores 90.0% on HumanEval+ compared to Codestral's 86.0%, while Claude Sonnet 4.5 scores 77.2% on SWE-bench Verified compared to Codestral's 40.0%, while Claude Sonnet 4.5 scores 62.0% on LiveCodeBench compared to Codestral's 48.0%.
Math: In math, Claude Sonnet 4.5 scores 87.0% on MATH compared to Codestral's 72.0%, while Claude Sonnet 4.5 scores 95.5% on GSM8K compared to Codestral's 88.0%.
Reasoning: In reasoning, Claude Sonnet 4.5 scores 83.4% on GPQA Diamond compared to Codestral's 44.0%, while Claude Sonnet 4.5 scores 45.0% on ARC-AGI compared to Codestral's 24.0%.
Context: In context, Claude Sonnet 4.5 scores 1.0M on Context Length compared to Codestral's 256K.
Pricing Comparison
Claude Sonnet 4.5 costs $3.0/1M input tokens and $15.0/1M output tokens, while Codestral costs $0.30/1M input and $0.90/1M output. Codestral is the more affordable option for API usage.
Speed Comparison
Claude Sonnet 4.5 generates output at 67 tok/s compared to Codestral's 90 tok/s, and the time to first token is 1170 ms for Claude Sonnet 4.5 versus 280 ms for Codestral. Codestral delivers faster throughput.
Verdict
For developers prioritizing coding and general intelligence and math, Claude Sonnet 4.5 has the edge. For those who value affordability and speed, Codestral is the stronger choice.
View Individual Model Pages
Claude Sonnet 4.5 vs Codestral — FAQ
Which is better, Claude Sonnet 4.5 or Codestral?
Claude Sonnet 4.5 wins on more benchmarks overall (10 vs 5). However, the best choice depends on your specific needs — each model excels in different areas.
How does Claude Sonnet 4.5 compare to Codestral for coding?
Claude Sonnet 4.5 is better for coding, scoring 77.2% on SWE-bench Verified compared to 40.0%. SWE-bench tests real-world software engineering by resolving actual GitHub issues.
Is Claude Sonnet 4.5 cheaper than Codestral?
Yes, Codestral is cheaper. Claude Sonnet 4.5 costs $3.0/1M input and $15.0/1M output tokens. Codestral costs $0.30/1M input and $0.90/1M output tokens.
Which is faster, Claude Sonnet 4.5 or Codestral?
Codestral is faster, generating output at 90 tok/s compared to 67 tok/s. Faster output speed means shorter wait times for API responses.
What benchmarks does the Claude Sonnet 4.5 vs Codestral comparison cover?
This comparison covers 15 benchmarks including Chatbot Arena ELO, MMLU-Pro, HumanEval+, MATH, GPQA Diamond, SWE-bench Verified, Output Speed, LiveCodeBench, and more. Metrics span general intelligence, coding, math, reasoning, speed, and cost categories.