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

Gemini 3.5×

GPT-5.5×

GPT-5×

DeepSeek-V4-Pro×Clear all

	Gemini 3.5 Google	GPT-5.5 OpenAI	GPT-5 OpenAI	DeepSeek-V4-Pro DeepSeek
Vendor	Google	OpenAI	OpenAI	DeepSeek
Family	Gemini	GPT	GPT	DeepSeek
Release date	2026-05-20	2026-04-23	2025-08-07	2026-04-22
Context window	2,000,000 tokens	400,000 tokens	400,000 tokens	1,000,000 tokens
Parameters	—	—	—	1.6T (49B active)
Modality	text, vision, audio, video	text, vision, audio	text, vision, audio	text
License	proprietary	proprietary	proprietary	MIT
Source	proprietary	proprietary	proprietary	open weights
Description	Google DeepMind's next-gen Gemini — positioned as "frontier intelligence with action". Built for complex agentic workflows. Announced at Google I/O 2026.	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.	OpenAI's flagship general-purpose model. Superseded by GPT-5.5 in April 2026.	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 →	vendor →	weights →
Benchmarks
MMLU-Pro	88.9%	87.6%	85.1%	84.2%
GPQA-D	91.2%	89.4%	87.3%	82.4%
HumanEval	—	96.8%	95.2%	95.1%
Aider	—	89.7%	85.4%	80.1%
AIME-25	96.1%	94.2%	91.0%	88.6%
LiveCB	—	78.9%	74.5%	72.4%
MMMU	86.0%	78.4%	74.1%	—