Eligibility — Models marked with supported open licenses/flags in seed data.
Primary ranking — GPQA Diamond, then MMLU, then newer release.
Variant collapse — We keep one row per model family (`familySlug` + parameter tier). When headline scores tie within ±0.5 pt (±10 Elo on Chatbot Arena), we pick the canonical SKU by lowest tracked input price, then GA over preview or limited access, then newest `release`. A folded sibling within the benchmark noise band can show a "Tied within margin" chip on that score cell.
Pricing — Hosted open-weight pricing still varies wildly — use the rate card column for apples-to-apples.

GPQA Diamond

#	Model	GPQA Diamond	Context	Input $/1M	Output $/1M
1	Kimi K2.6 ReasoningVisionTools GPQA Diamond: 90.5%	90.5%	262K	$0.75	$3.50
2	DeepSeek V4 Pro ReasoningTools GPQA Diamond: 90.1%	90.1%	1M	$0.43	$0.87
3	Qwen3.5-397B-A17B ReasoningTools GPQA Diamond: 89.3%	89.3%	262K	$0.39	$2.34
4	Trinity-Large-Thinking ReasoningTools GPQA Diamond: 89.2%	89.2%	256K	$0.22	$0.85
5	DeepSeek V4 Flash ReasoningTools GPQA Diamond: 88.1%	88.1%	1M	$0.14	$0.28
6	Kimi K2.5 Tools GPQA Diamond: 87.9%	87.9%	256K	$0.44	$2.00
7	Qwen3.6-27B ReasoningVisionTools GPQA Diamond: 87.8%	87.8%	262K	$0.32	$3.20
8	GLM-5.1 ReasoningTools GPQA Diamond: 86.2%	86.2%	200k	$1.05	$3.50
9	Qwen3-235B-A22B GPQA Diamond: 86.1%	86.1%	128K	$0.40	$1.20
10	Qwen3.6-35B-A3B Tools GPQA Diamond: 86%	86%	262K	$0.15	$1.00
11	Qwen3.5-27B ReasoningVisionTools GPQA Diamond: 85.8%	85.8%	262K	$0.20	$1.56
12	Gemma 4 31B Tools GPQA Diamond: 85.7%Tied within margin	85.7%Tied within margin	256k	—	—
13	Qwen3.5-122B-A10B ReasoningVisionTools GPQA Diamond: 85.7%	85.7%	262K	$0.26	$2.08
14	Qwen3.5-35B-A3B ReasoningTools GPQA Diamond: 84.5%	84.5%	262K	$0.16	$1.30
15	DeepSeek V3.2 GPQA Diamond: 84%	84%	160K	$0.25	$0.38
16	Qwen3.5-9B VisionTools GPQA Diamond: 81.7%	81.7%	262K	$0.10	$0.15
17	DeepSeek R1 0528 Reasoning GPQA Diamond: 81%	81%	160K	$0.10	$0.30
18	Gemma 4 26B A4B IT Tools GPQA Diamond: 79.2%	79.2%	256k	$0.06	$0.33
19	K-EXAONE 236B-A23B GPQA Diamond: 78.3%	78.3%	256k	—	—
20	gpt-oss-120b Tools GPQA Diamond: 78.2%	78.2%	131K	$0.04	$0.18

Honorable mentions

Next seats in this ranking. Lines below are from each model's stored description in LLMReference seed data—spot-check the model page before relying on a capability claim.

#4Trinity-Large-Thinking
Arcee AI's flagship 400B sparse MoE reasoning model with 13B active parameters per token. Trained on 20T tokens with a STEM-focused curriculum. Designed for agentic workflows, chain-of-thought reasoning, and long-context tasks up to 256K tokens (BF16 API). Open-source under Apache 2.0. Available via Arcee AI API.
89.2%
GPQA Diamond
#5DeepSeek V4 Flash
DeepSeek V4 Flash is a 284B parameter (13B activated) Mixture-of-Experts language model with 1M-token context. Features a hybrid attention architecture combining Compressed Sparse Attention (CSA) and Heavily Compressed Attention (HCA) for efficient long-context inference. Supports thinking and non-thinking modes. Legacy API aliases deepseek-chat and deepseek-reasoner map to this model's non-thinking and thinking modes respectively. Pricing: $0.14/1M input, $0.28/1M output (cache hit: $0.0028/1M input). MIT licensed.
88.1%
GPQA Diamond
#6Kimi K2.5
Moonshot Kimi K2.5 available on AWS Bedrock
87.9%
GPQA Diamond

Compare Top Picks

Kimi K2.6 vs DeepSeek V4 Pro Kimi K2.6 vs Qwen3.5-397B-A17B Kimi K2.6 vs Trinity-Large-Thinking Kimi K2.6 vs DeepSeek V4 Flash DeepSeek V4 Pro vs Qwen3.5-397B-A17B DeepSeek V4 Pro vs Trinity-Large-Thinking

Browse Other Categories

Best LLMs for Code Generation Best LLMs for RAG Best AI Agents & Agentic Models Best LLMs for Classification Best Multimodal / Vision LLMs Best LLMs for Reasoning & Math Best Small Language Models (SLMs)Best LLMs for Function Calling & Tool Use Cheapest LLM APIs You Can Call Right Now Best Long Context LLMs Best Mainstream LLM APIs, Ranked Best LLMs for Enterprise Best Free LLMs You Can Use Right Now Best LLMs for Writing Best LLMs for Marketing Best LLMs for Customer Support