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

Mistral Large 2×

DeepSeek-V4-Pro×

Gemini 3.5×

Gemini Omni×Clear all

	Mistral Large 2 Mistral	DeepSeek-V4-Pro DeepSeek	Gemini 3.5 Google	Gemini Omni Google
Vendor	Mistral	DeepSeek	Google	Google
Family	Mistral	DeepSeek	Gemini	Gemini
Release date	2024-07-24	2026-04-22	2026-05-20	2026-05-20
Context window	128,000 tokens	1,000,000 tokens	2,000,000 tokens	1,000,000 tokens
Parameters	123B	1.6T (49B active)	—	—
Modality	text	text	text, vision, audio, video	text, vision, audio, video
License	Mistral Research License	MIT	proprietary	proprietary
Source	open weights	open weights	proprietary	proprietary
Description	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.	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.	Google DeepMind's next-gen Gemini — positioned as "frontier intelligence with action". Built for complex agentic workflows. Announced at Google I/O 2026.	Multimodal Gemini variant introduced at Google I/O 2026 — unified text, image, audio, and video processing in a single model.
Links	vendor →weights →	weights →	vendor →	vendor →
Benchmarks
MMLU-Pro	—	84.2%	88.9%	—
GPQA-D	—	82.4%	91.2%	—
HumanEval	89.0%	95.1%	—	—
Aider	—	80.1%	—	—
AIME-25	—	88.6%	96.1%	—
LiveCB	—	72.4%	—	—
MMMU	—	—	86.0%	82.5%