prices synced 2026-09-11
A

Olmo 3 32B Think

AllenAI · Released Nov 2025

Compare

Olmo 3 32B Think is a 32-billion-parameter language model from Allen Institute for AI with an extended reasoning capability and 65k token context window.

Strengths
Shows solid performance on complex reasoning tasks through its think mode, which allows the model to work through problems step-by-step before responding.
Best for
Problem-solving and analytical workloads where intermediate reasoning steps are valuable, particularly those fitting within a 65k token context.
Limitations
The 32B parameter size means it trades some general capability for a smaller model footprint; extended reasoning adds latency compared to direct-response models.

Input / 1M

$0.15

Output / 1M

$0.5

Cached input / 1M

Context window

65K

Price history

Price per 1M tokens over timeOutput $0.5; Input $0.15 as of Jun 2026.$0$0.1$0.2$0.3$0.4$0.5Output on 11 Jun 2026: $0.5Out $0.5Input on 11 Jun 2026: $0.15In $0.15Jun 2026

Snapshots

Effective Input Output Cached in Note Source
11 Jun 2026 $0.15 $0.5 Imported from OpenRouter openrouter.ai
Report a problem