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×

Claude 4.8 Opus×

DeepSeek-V3.1×

GPT-5.5×Clear all

	o3 OpenAI	Claude 4.8 Opus Anthropic	DeepSeek-V3.1 DeepSeek	GPT-5.5 OpenAI
Vendor	OpenAI	Anthropic	DeepSeek	OpenAI
Family	o-series	Claude	DeepSeek	GPT
Release date	2025-04-16	2026-05-28	2025-08-21	2026-04-23
Context window	200,000 tokens	1,000,000 tokens	128,000 tokens	400,000 tokens
Parameters	—	—	671B	—
Modality	text, vision	text, vision	text	text, vision, audio
License	proprietary	proprietary	MIT	proprietary
Source	proprietary	proprietary	open weights	proprietary
Description	Reasoning-focused model in the o-series.	Anthropic's flagship hybrid-reasoning model — Opus 4.8 pushes coding and AI-agent workflows further (agentic coding 69.2% vs 64.3% on 4.7; multidisciplinary reasoning 57.9% vs 54.7%). 1M-token context, $5/$25 per MTok pricing held flat from Opus 4.7. Introduces Dynamic Workflows for Claude Code (research preview). GA on Anthropic API (`claude-opus-4-8`), AWS Bedrock, Google Vertex, Microsoft Foundry, and GitHub Copilot.	Large MoE open-weight model. Predecessor to DeepSeek-V4.	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.
Links		vendor →	weights →	vendor →
Benchmarks
MMLU-Pro	—	—	—	87.6%
GPQA-D	87.7%	—	—	89.4%
HumanEval	—	—	—	96.8%
Aider	—	—	—	89.7%
AIME-25	88.9%	—	—	94.2%
LiveCB	—	—	—	78.9%
MMMU	—	—	—	78.4%