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×

DeepSeek-V4-Flash×

GPT-5.5×

Mistral Large 2×Clear all

	GPT-5 OpenAI	DeepSeek-V4-Flash DeepSeek	GPT-5.5 OpenAI	Mistral Large 2 Mistral
Vendor	OpenAI	DeepSeek	OpenAI	Mistral
Family	GPT	DeepSeek	GPT	Mistral
Release date	2025-08-07	2026-04-22	2026-04-23	2024-07-24
Context window	400,000 tokens	1,000,000 tokens	400,000 tokens	128,000 tokens
Parameters	—	284B (13B active)	—	123B
Modality	text, vision, audio	text	text, vision, audio	text
License	proprietary	MIT	proprietary	Mistral Research License
Source	proprietary	open weights	proprietary	open weights
Description	OpenAI's flagship general-purpose model. Superseded by GPT-5.5 in April 2026.	Smaller, faster sibling to DeepSeek-V4-Pro. Same 1M context window with a much lighter 284B / 13B-active MoE.	OpenAI's smartest and most intuitive model — successor to GPT-5, with major coding, research, and document workflow gains. GPT-5.5 Pro variant also available for heavier reasoning.	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	vendor →	weights →	vendor →	vendor →weights →
Benchmarks
MMLU-Pro	85.1%	—	87.6%	—
GPQA-D	87.3%	—	89.4%	—
HumanEval	95.2%	—	96.8%	89.0%
Aider	85.4%	—	89.7%	—
AIME-25	91.0%	—	94.2%	—
LiveCB	74.5%	—	78.9%	—
MMMU	74.1%	—	78.4%	—