modelgrep

Anthropic: Claude 3.5 Haiku vs DeepSeek: R1

DeepSeek: R1 wins on more metrics (6 of 8), but the right pick depends on what you optimize for — see the breakdown below.

MetricAnthropic: Claude 3.5 HaikuDeepSeek: R1
Intelligence Index18.718.8
Coding Index10.715.9
GPQA Diamond41%71%
Design Arena Elo
Speed (tokens/sec)4072
Latency755ms1.4s
Input price /M$0.800$0.700
Output price /M$4.00$2.50
Context window200K164K
CapabilitiesToolsVisionReasoningToolsJSON