← All comparisons

DeepSeek V4 Flash (Reasoning, Max Effort) vs Claude 3.5 Sonnet (June '24)

DeepSeek vs Anthropic — side-by-side benchmark comparison

DeepSeek V4 Flash (Reasoning, Max Effort)Claude 3.5 Sonnet (June '24)
Intelligence Index46.514.2
Coding Index38.726.0
Math Index
Output speed (tok/s)119.30.0
Blended price ($/1M)$0.17$6.56
Time to first token (s)0.86s0.00s
aime9.7%
aime 25
artificial analysis coding index38.7026.00
artificial analysis intelligence index46.5014.20
artificial analysis math index
gpqa89.4%56.0%
hle32.1%3.7%
ifbench79.2%
lcr63.0%
livecodebench
math 50069.5%
mmlu pro75.1%
scicode44.9%31.6%
tau295.0%
terminalbench hard35.6%

Benchmark data from Artificial Analysis.