Compare models

Pick up to 4models. Specs render side-by-side. Share the URL — it's stateless.

Comparing agents instead? Switch to agent compare →

Selected

o3×

DeepSeek-V4-Pro×

Grok 4×

Mistral Large 2×Clear all

	o3 OpenAI	DeepSeek-V4-Pro DeepSeek	Grok 4 xAI	Mistral Large 2 Mistral
Vendor	OpenAI	DeepSeek	xAI	Mistral
Family	o-series	DeepSeek	Grok	Mistral
Release date	2025-04-16	2026-04-22	2025-07-09	2024-07-24
Context window	200,000 tokens	1,000,000 tokens	256,000 tokens	128,000 tokens
Parameters	—	1.6T (49B active)	—	123B
Modality	text, vision	text	text	text
License	proprietary	MIT	proprietary	Mistral Research License
Source	proprietary	open weights	proprietary	open weights
Description	Reasoning-focused model in the o-series.	DeepSeek's flagship open-weight MoE. 1.6T parameters with 49B activated, 1M-token context, and a hybrid attention scheme (CSA + HCA) that delivers long-context inference at ~27% of V3.2's FLOPs.	—	Large text-only Mistral model with a 128K context window and 123B parameters, tuned for strong instruction following and long-context reasoning. Mistral's flagship open-weights release in the Large 2 line.
Links		weights →	vendor →	vendor →weights →
Benchmarks
MMLU-Pro	—	84.2%	82.8%	—
GPQA-D	87.7%	82.4%	87.5%	—
HumanEval	—	95.1%	—	89.0%
Aider	—	80.1%	72.5%	—
AIME-25	88.9%	88.6%	86.1%	—
LiveCB	—	72.4%	65.3%	—