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×

Gemini 2.5 Flash×

DeepSeek-V4-Pro×

Qwen3 235B×Clear all

	Qwen3.7-Max Alibaba	Gemini 2.5 Flash Google	DeepSeek-V4-Pro DeepSeek	Qwen3 235B Alibaba
Vendor	Alibaba	Google	DeepSeek	Alibaba
Family	Qwen	Gemini	DeepSeek	Qwen
Release date	2026-05-20	2025-06-17	2026-04-22	2025-04-29
Context window	1,000,000 tokens	1,000,000 tokens	1,000,000 tokens	128,000 tokens
Parameters	—	—	1.6T (49B active)	235B
Modality	text	text, vision, audio	text	text
License	proprietary	proprietary	MIT	Apache-2.0
Source	proprietary	proprietary	open weights	open weights
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.	Google's cost-and-latency-optimized Gemini 2.5 variant. Generally available June 2025 — designed for high-throughput agentic workflows with sub-second first-token latency.	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.	Predecessor to the Qwen3.6 family.
Links	vendor →		weights →	weights →
Benchmarks
MMLU-Pro	83.7%	—	84.2%	—
GPQA-D	83.0%	—	82.4%	—
HumanEval	93.9%	—	95.1%	—
Aider	78.4%	—	80.1%	—
AIME-25	90.4%	—	88.6%	—
LiveCB	71.0%	—	72.4%	—