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

GPT-5×

Mistral Large 2×

DeepSeek-V4-Pro×

Grok 4×Clear all

	GPT-5 OpenAI	Mistral Large 2 Mistral	DeepSeek-V4-Pro DeepSeek	Grok 4 xAI
Vendor	OpenAI	Mistral	DeepSeek	xAI
Family	GPT	Mistral	DeepSeek	Grok
Release date	2025-08-07	2024-07-24	2026-04-22	2025-07-09
Context window	400,000 tokens	128,000 tokens	1,000,000 tokens	256,000 tokens
Parameters	—	123B	1.6T (49B active)	—
Modality	text, vision, audio	text	text	text
License	proprietary	Mistral Research License	MIT	proprietary
Source	proprietary	open weights	open weights	proprietary
Description	OpenAI's flagship general-purpose model. Superseded by GPT-5.5 in April 2026.	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.	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.	—
Links	vendor →	vendor →weights →	weights →	vendor →
Benchmarks
MMLU-Pro	85.1%	—	84.2%	82.8%
GPQA-D	87.3%	—	82.4%	87.5%
HumanEval	95.2%	89.0%	95.1%	—
Aider	85.4%	—	80.1%	72.5%
AIME-25	91.0%	—	88.6%	86.1%
LiveCB	74.5%	—	72.4%	65.3%
MMMU	74.1%	—	—	—