modelgrep

AllenAI: Olmo 3 32B Think vs Meta: Llama 3.1 70B Instruct

Meta: Llama 3.1 70B Instruct wins on more metrics (6 of 8), but the right pick depends on what you optimize for — see the breakdown below.

MetricAllenAI: Olmo 3 32B ThinkMeta: Llama 3.1 70B Instruct
Intelligence Index12.112.5
Coding Index10.510.9
GPQA Diamond61%41%
Design Arena Elo
Speed (tokens/sec)29
Latency290ms
Input price /M$0.150$0.400
Output price /M$0.500$0.400
Context window66K131K
CapabilitiesReasoningJSONToolsJSON