Leaderboard

AI Model Leaderboard

28 models compared side by side: benchmark scores, context windows, pricing, and speed. Every number comes from a public source or provider announcement -- no guesses, no missing data marked as zero. Filter by provider, task, or license. Updated as new models ship.

0

Models tracked

0

Providers

0

With benchmarks

0

Open source

Model Rankings

Top models ranked by publicly reported benchmark score. Every number links to its source.

Top models per task

Best-in-class per task type, ranked by publicly reported benchmark scores. Rows with no verified score are omitted.

Coding

SWE-bench Pro

Reasoning

GPQA Diamond

Agents

OSWorld

Vision

MMMU

Long Context

Context window

Budget

Blended price per MTok, lower is better

Benchmark leaderboards

Top models per benchmark. Click any row to see full model specs.

Fastest and most affordable models

Median output speed as measured by third-party benchmarks (Artificial Analysis / OpenRouter); pricing as listed by the provider.

Fastest (tokens per second)

ModelProviderTok/sIn $/MTok
Gemini 2.5 FlashGoogle222$0.30
Qwen3.7 MaxAlibaba206$1.25
Gemini 3.5 FlashGoogle198$1.50
o3OpenAI167$2.00
Mistral Small 3Mistral166$0.10
GPT-5.4OpenAI165$2.50
o4-miniOpenAI163$1.10

Most affordable (input cost)

ModelProviderIn $/MTokOut $/MTok
Llama 4 ScoutMeta$0.08$0.30
DeepSeek V4 FlashDeepSeek$0.09$0.18
Mistral Small 3Mistral$0.10$0.30
Llama 4 MaverickMeta$0.15$0.60
Tencent Hy3Tencent$0.20$0.80
Gemini 2.5 FlashGoogle$0.30$2.50
DeepSeek V4 ProDeepSeek$0.43$0.87

Model comparison

28 of 28 models. Sort by any column. Filter by provider, license, or task.

Model Provider OpenContext In $/MTok Out $/MTokTok/s Released Best for
Gemini 3.5 ProGoogle2.1M------2026-07
Long CtxReasoningVision
Tencent Hy3Tencent262K$0.20$0.80--2026-07
CodeReasoningAgents
Claude Sonnet 5Anthropic1.0M$2.00$10.00792026-06
CodeAgentsReasoning
Sakana Fugu UltraSakana AI1.0M$5.00$30.00--2026-06
ReasoningAgentsCode
GLM-5.2Z.ai1.0M$0.90$2.86--2026-06
CodeReasoningAgents
Claude Fable 5Anthropic1.0M$10.00$50.00702026-06
CodeAgentsReasoning
Claude Opus 4.8Anthropic1.0M$5.00$25.00652026-05
CodeAgentsReasoning
Gemini 3.5 FlashGoogle1.0M$1.50$9.001982026-05
CodeAgentsBudget
DeepSeek V4 ProDeepSeek1.0M$0.43$0.87722026-04
CodeReasoningBudget
DeepSeek V4 FlashDeepSeek1.0M$0.09$0.181132026-04
BudgetCode
GPT-5.5OpenAI1.0M$5.00$30.00892026-04
CodeReasoningAgents
Qwen3.7 MaxAlibaba1.0M$1.25$3.752062026-04
ReasoningCodeLong Ctx
Grok 4.20xAI2.0M$2.00$6.00--2026-03
ReasoningAgentsLong Ctx
GPT-5.4OpenAI1.0M$2.50$15.001652026-03
CodeReasoningVision
Claude Sonnet 4.6Anthropic1.0M$3.00$15.00472026-02
CodeAgentsVision
Claude Haiku 4.5Anthropic200K$1.00$5.00982025-10
BudgetAgents
Mistral Small 3Mistral128K$0.10$0.301662025-09
BudgetCode
GPT-5OpenAI400K$0.63$5.001072025-08
ReasoningCode
Gemini 2.5 ProGoogle1.0M$1.25$10.001512025-06
Long CtxReasoningVision
Gemini 2.5 FlashGoogle1.0M$0.30$2.502222025-05
BudgetCodeAgents
Qwen3 MaxAlibaba262K$0.78$3.90662025-04
ReasoningCode
o4-miniOpenAI200K$1.10$4.401632025-04
ReasoningCodeBudget
o3OpenAI200K$2.00$8.001672025-04
ReasoningCode
Llama 4 MaverickMeta1.0M$0.15$0.601182025-04
VisionReasoningBudget
Llama 4 ScoutMeta10.0M$0.08$0.301092025-04
Long CtxBudgetVision
Grok 3xAI131K$3.00$15.00--2025-02
ReasoningCode
DeepSeek R1DeepSeek128K$0.55$2.19--2025-01
ReasoningCode
Mistral Large 2Mistral128K$2.00$6.00552024-07
CodeReasoning

Context window, cost and speed

Full specs side by side. Click a column header to sort.

Model ProviderContext Max outputIn $/MTok Out $/MTokTok/s Pricing note
Llama 4 ScoutMeta10.0M--$0.08$0.30109Via third-party inference providers; Meta does not operate a paid API
Gemini 3.5 ProGoogle2.1M--------GA expected Jul 17 2026; pricing not confirmed as of 2026-07-06
Grok 4.20xAI2.0M--$2.00$6.00--
Claude Fable 5Anthropic1.0M128K$10.00$50.0070
Claude Sonnet 5Anthropic1.0M128K$2.00$10.0079Introductory pricing through Aug 31 2026; standard rate $3/$15 per MTok
Claude Opus 4.8Anthropic1.0M128K$5.00$25.0065
Claude Sonnet 4.6Anthropic1.0M128K$3.00$15.0047
GPT-5.5OpenAI1.0M--$5.00$30.0089
GPT-5.4OpenAI1.0M128K$2.50$15.00165
Gemini 3.5 FlashGoogle1.0M66K$1.50$9.00198
Gemini 2.5 ProGoogle1.0M--$1.25$10.00151$2.50/$15 per MTok for prompts over 200K tokens
Gemini 2.5 FlashGoogle1.0M66K$0.30$2.50222
GLM-5.2Z.ai1.0M131K$0.90$2.86--Via OpenRouter (model ID z-ai/glm-5.2); SiliconFlow lists $1.40/$4.40 per MTok
Llama 4 MaverickMeta1.0M--$0.15$0.60118Via third-party inference providers; Meta does not operate a paid API
DeepSeek V4 ProDeepSeek1.0M384K$0.43$0.877275% launch promo made permanent 2026-05-22 (original list was $1.74/$3.48 per MTok); $0.435 is now the standard rate.
DeepSeek V4 FlashDeepSeek1.0M384K$0.09$0.18113Cached input $0.0028/MTok
Qwen3.7 MaxAlibaba1.0M--$1.25$3.75206
Sakana Fugu UltraSakana AI1.0M131K$5.00$30.00--$10/$45 per MTok for contexts over 272K tokens; cached input $0.50/MTok (standard), $1.00/MTok (over 272K)
GPT-5OpenAI400K--$0.63$5.00107Succeeded by GPT-5.5 (official OpenAI replacement); no longer on pricing page. API shutdown Dec 11 2026.
Qwen3 MaxAlibaba262K--$0.78$3.9066
Tencent Hy3Tencent262K--$0.20$0.80--Free via OpenRouter for ~2 weeks from 2026-07-07 (model ID tencent/hy3:free); paid tier $0.20/$0.80 per MTok via OpenRouter
Claude Haiku 4.5Anthropic200K--$1.00$5.0098
o4-miniOpenAI200K--$1.10$4.40163
o3OpenAI200K--$2.00$8.00167
Grok 3xAI131K--$3.00$15.00--
DeepSeek R1DeepSeek128K--$0.55$2.19--deepseek-reasoner API alias now routes to V4 Flash thinking mode
Mistral Large 2Mistral128K--$2.00$6.0055
Mistral Small 3Mistral128K--$0.10$0.30166

Benchmark glossary

What each benchmark actually measures and why it matters.

SWE-bench Verified

Real GitHub issues from popular Python repos. The model must write code that passes the existing test suite. "Verified" means a human checked that each issue is solvable. Scores range 0-100%.

swebench.com

MMLU

Massive Multitask Language Understanding. 57 academic subjects from high-school to professional level (law, medicine, STEM). Tests breadth of knowledge. Scores are % correct.

Hendrycks et al.

HumanEval

OpenAI benchmark: 164 hand-crafted Python programming problems. Measures functional code generation (pass@1). Widely used but considered dated for frontier models.

OpenAI

MMMU

Massive Multidisciplinary Multimodal Understanding. 11.5K questions requiring images (charts, diagrams, photos) plus text. Standard for vision-language model evaluation.

mmmu-benchmark.github.io

RULER

Ruler for Long-Context Evaluation. Tests retrieval, multi-hop reasoning, and aggregation over very long documents (up to 128K tokens). Better signal than "needle in haystack" for long-context claims.

Hsieh et al.

GPQA Diamond

Graduate-Level Google-Proof Q&A, Diamond split. Expert-level science questions that Google cannot directly answer. High signal for deep scientific reasoning. Human experts score ~65%.

Rein et al.

MATH

Competition mathematics (AMC, AIME, Olympiad). 12,500 problems across 7 difficulty levels. Tests symbolic reasoning and multi-step derivation. Scores are % correct.

Hendrycks et al.

Not sure which model fits your project?

We evaluate models against your actual use case, run cost-to-quality tests across providers, and build the integration. No vendor lock-in, no guesswork.

Related guides

How we verify. Benchmark scores are taken from official provider pages, third-party leaderboards, or peer-reviewed papers. Where a number could not be independently confirmed it is shown as a dash. Pricing reflects the public API rate at the time of last update; check the provider for current pricing.

Affiliate disclosure. AY Automate has no affiliate relationship with any model provider listed here. Rankings are editorial, not commercial.