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

Qwen3.7-Max×

DeepSeek-V3.1×

DeepSeek-V4-Flash×

Gemini 3.5×Clear all

	Qwen3.7-Max Alibaba	DeepSeek-V3.1 DeepSeek	DeepSeek-V4-Flash DeepSeek	Gemini 3.5 Google
Vendor	Alibaba	DeepSeek	DeepSeek	Google
Family	Qwen	DeepSeek	DeepSeek	Gemini
Release date	2026-05-20	2025-08-21	2026-04-22	2026-05-20
Context window	1,000,000 tokens	128,000 tokens	1,000,000 tokens	2,000,000 tokens
Parameters	—	671B	284B (13B active)	—
Modality	text	text	text	text, vision, audio, video
License	proprietary	MIT	MIT	proprietary
Source	proprietary	open weights	open weights	proprietary
Description	Alibaba's flagship agent model — 1M-token context, extended-thinking mode, 56.6 on the Artificial Analysis Intelligence Index v4.0 (5th overall, #1 Chinese). 50.8% on Terminal-Bench Hard. Designed for long-horizon agent workloads (hundreds-to-thousands of steps). Closed-weight, $2.50/$7.50 per 1M tokens.	Large MoE open-weight model. Predecessor to DeepSeek-V4.	Smaller, faster sibling to DeepSeek-V4-Pro. Same 1M context window with a much lighter 284B / 13B-active MoE.	Google DeepMind's next-gen Gemini — positioned as "frontier intelligence with action". Built for complex agentic workflows. Announced at Google I/O 2026.
Links	vendor →	weights →	weights →	vendor →
Benchmarks
MMLU-Pro	83.7%	—	—	88.9%
GPQA-D	83.0%	—	—	91.2%
HumanEval	93.9%	—	—	—
Aider	78.4%	—	—	—
AIME-25	90.4%	—	—	96.1%
LiveCB	71.0%	—	—	—
MMMU	—	—	—	86.0%