← All comparisons

Claude 4.5 Haiku (Non-reasoning) vs Mistral Medium 3.1

Anthropic vs Mistral — side-by-side benchmark comparison

Claude 4.5 Haiku (Non-reasoning)Mistral Medium 3.1
Intelligence Index31.021.3
Coding Index29.618.3
Math Index39.038.3
Output speed (tok/s)105.864.0
Blended price ($/1M)$2.19$0.80
Time to first token (s)0.65s0.52s
aime
aime 2539.0%38.3%
artificial analysis coding index29.6018.30
artificial analysis intelligence index31.0021.30
artificial analysis math index39.0038.30
gpqa64.6%58.8%
hle4.3%4.4%
ifbench42.0%39.8%
lcr43.7%19.7%
livecodebench51.1%40.6%
math 500
mmlu pro80.0%68.3%
scicode34.4%33.8%
tau232.5%40.6%
terminalbench hard27.3%10.6%

Benchmark data from Artificial Analysis.