LIVETracking 1,072 AI models from 273 providers.
BenchGeckoBeta
AI
Providers
Economy
Mindshare
Infra
Learn
Labs
Developers
Charts·Build live AI market viewsOpen chartsBuild your own chart
Home·Models
BenchGecko

BenchGecko is the data layer of the AI economy. Thousands of models with cross-provider pricing and daily price history. Hundreds of companies with valuations, funding timelines, and revenue estimates. Benchmark scores, developer adoption signals, agent leaderboards, and a changelog that captures every price drop, every launch, every deprecation as it happens. If it moved in AI today, it's already on BenchGecko.

@BenchGecko

Rankings

  • Leaderboard
  • All Models
  • AI Agents
  • Benchmarks
  • Compare Models
  • MCP Servers
  • Skills
  • Pricing

Top Models

  • DeepSeek V3.1 Terminus
  • Qwen3 VL 235B A22B Instruct
  • Qwen3 VL 235B A22B Thinking
  • Trinity Large Preview (free)
  • GLM 4.5 Air
  • GLM 4.7 Flash
  • MiniMax M1
  • GLM 4.5V
  • Gemma 3 12B
  • Nova 2 Lite

Providers

  • ~anthropic
  • ~google
  • ~moonshotai
  • ~openai
  • AI Research
  • AI21 Labs
  • AI4Bharat
  • aion-labs
  • All Providers →

Benchmarks

  • Aider · Code Editing
  • Aider polyglot
  • ANLI
  • APEX-Agents
  • ARC AI2
  • ARC-AGI
  • ARC-AGI-2
  • Artificial Analysis · Agentic Index
  • All Benchmarks →

Resources

  • API Documentation
  • Methodology
  • Research
  • Gecko Mindshare
  • News
  • Changelog
  • Countries
  • Model Status
  • Brand Assets
  • Developers
  • Charts
  • Learn
  • Economy
  • Labs
  • Gecko Tests
© 2026 BenchGecko·The AI Economy, Tracked
Privacy·Terms·Press
Models · LeaderboardUpdated 30d ago · 1,201 models · 273 providers · 130 benchmarks

Every AI Model · Tracked

The most complete list of AI models you can actually use · 1,201 models, 273 providers, 130 benchmarks · all scored, priced, and ranked in one place.

Benchmark scoredCross-provider pricingContext · parameters · licensingRefreshed daily
Compare side by sideShowing 159 of 1,201
Total Models
1,201
Providers
273
Top Model
GPT-5.5 Pro
87.8%
Cheapest Capable
gpt-oss-20b
$0.03/M
PodiumMoversLeaderboardFAQ

Top 10 Overall

Ranked by average benchmark score · min 3 benchmarks

#1 GOLD3 benchmarks
OpenAI logo
GPT-5.5 Pro
OpenAI
87.8%
#2 SILVER6 benchmarks
OpenAI logo
GPT-5.5
OpenAI
85.0%
#3 BRONZE7 benchmarks
OpenAI logo
GPT-5 Chat
OpenAI
81.9%
#4Anthropic logo
Claude Mythos Preview
Anthropic
81.8%14 benchmarks
#5Alibaba Qwen logo
Qwen3.5 397B A17B
Alibaba Qwen
78.4%11 benchmarks
#6DeepSeek logo
DeepSeek V3.2 Speciale
DeepSeek
78.2%9 benchmarks
#7Anthropic logo
Claude Instant
Anthropic
78.0%4 benchmarks
#8stepfun logo
Step 3.5 Flash
stepfun
76.9%10 benchmarks
#9DeepSeek logo
DeepSeek-V2 (MoE-236B, May 2024)
DeepSeek
76.5%7 benchmarks
#10z-ai logo
GLM 5.2
z-ai
76.2%13 benchmarks

What's moving

New releases, coverage, price leaders, and ELO champions

#ModelProviderCategoryReleased
1DeepSeek logoDeepSeek V4.1 FlashDeepSeekMultimodalSep 10, 2026
2nex-agi logoNex-N2.5-Mini (free)nex-agiLLMSep 8, 2026
3inception logoMercury 2.5inceptionLLMSep 8, 2026
4nex-agi logoNex-N2.5-Pro (free)nex-agiMultimodalSep 8, 2026
5OpenAI logoGPT-6 AstraOpenAIMultimodalSep 4, 2026
6OpenAI logoGPT-6 Astra (batch)OpenAIMultimodalSep 4, 2026
7OpenAI logoGPT-6 Astra ProOpenAIMultimodalSep 4, 2026
8OpenAI logoGPT-6 Astra Pro (batch)OpenAIMultimodalSep 4, 2026
9Alibaba Qwen logoQwen3.8 Max (0902)Alibaba QwenMultimodalSep 3, 2026
10Google DeepMind logoGemini 3.8 Flash (batch)Google DeepMindMultimodalSep 2, 2026
Provider
Filter
159 results

All Models

#ModelProviderCategoryContextIn $/MOut $/MAvgBenchmarksELO
1Anthropic logoClaude Mythos PreviewAnthropicLLM1.0MTBDTBD81.8%14—
2Google DeepMind logoGemini 2.5 Pro Preview 05-06Google DeepMindMultimodal1.0M$1.25$10.0076.9%1—
3z-ai logoGLM 5.2OSSz-aiLLM1.0M$1.00$4.0076.2%13—
4Anthropic logoClaude Fable 5AnthropicMultimodal1.0M$10.00$50.0074.4%13—
5DeepSeek logoDeepSeek V4 ProOSSDeepSeekLLM1.0M$0.96$1.9173.6%12—
6minimax logoMiniMax M3OSSminimaxMultimodal1.0M$0.30$1.2070.0%13—
7writer logoPalmyra X5writerLLM1.0M$0.60$6.0069.7%5—
8Alibaba Qwen logoQwen3.7 MaxOSSAlibaba QwenLLM1.0M$1.25$3.7568.1%18—
9DeepSeek logoDeepSeek V4 FlashOSSDeepSeekLLM1.0M$0.09$0.1867.2%12—
10xAI logoGrok 4.3xAIMultimodal1.0M$1.25$2.5066.7%13—
11xAI logoGrok 4.20xAIMultimodal2.0M$1.25$2.5066.1%4—
12Google DeepMind logoGemini 2.0 Flash LiteGoogle DeepMindMultimodal1.0M$0.07$0.3064.2%5—
13OpenAI logoGPT-5.4 ProOpenAIMultimodal1.1M$30.00$180.0063.2%10—
14Google DeepMind logoGemini 3.5 FlashGoogle DeepMindMultimodal1.0M$1.50$9.0062.6%19—
15Anthropic logoClaude Opus 4.8AnthropicMultimodal1.0M$5.00$25.0061.4%16—
16Alibaba Qwen logoQwen3.6 PlusOSSAlibaba QwenMultimodal1.0M$0.33$1.9559.9%20—
17Google DeepMind logoGemini 2.5 Flash LiteGoogle DeepMindMultimodal1.0M$0.10$0.4059.1%8—
18OpenAI logoGPT-5.4OpenAIMultimodal1.1M$2.50$15.0058.9%22—
19Anthropic logoClaude Opus 4.7AnthropicMultimodal1.0M$5.00$25.0058.2%20—
20xiaomi logoMiMo-V2-ProxiaomiLLM1.0M$1.00$3.0058.1%13—
21Anthropic logoClaude Opus 4.6AnthropicMultimodal1.0M$5.00$25.0057.1%21—
22Google DeepMind logoGemini 3.1 Pro PreviewGoogle DeepMindMultimodal1.0M$2.00$12.0056.6%28—
23Google DeepMind logoGemini 2.5 ProGoogle DeepMindMultimodal1.0M$1.25$10.0052.7%45—
24NVIDIA logoNemotron 3 UltraOSSNVIDIALLM1.0M$0.50$2.2051.8%11—
25Google DeepMind logoGemini 2.5 Pro Preview 06-05Google DeepMindMultimodal1.0M$1.25$10.0050.9%4—
26Alibaba Qwen logoQwen3.6 FlashOSSAlibaba QwenMultimodal1.0M$0.19$1.1349.8%14—
27Google DeepMind logoGemini 2.0 FlashGoogle DeepMindMultimodal1.0M$0.10$0.4048.0%20—
28Anthropic logoClaude Sonnet 4.6AnthropicMultimodal1.0M$3.00$15.0048.0%21—
29Google DeepMind logoGemini 3 Flash PreviewGoogle DeepMindMultimodal1.0M$0.50$3.0047.7%26—
30OpenAI logoGPT-4.1 MiniOpenAIMultimodal1.0M$0.40$1.6044.5%16—
31Anthropic logoClaude Opus 4.6 (Fast)AnthropicMultimodal1.0M$30.00$150.0043.4%12—
32Anthropic logoClaude Sonnet 4AnthropicMultimodal1.0M$3.00$15.0043.3%28—
33OpenAI logoGPT-4.1OpenAIMultimodal1.0M$2.00$8.0043.3%22—
34Anthropic logoClaude Sonnet 4.5AnthropicMultimodal1.0M$3.00$15.0039.6%23—
35Google DeepMind logoGemini 2.5 FlashGoogle DeepMindMultimodal1.0M$0.30$2.5038.5%26—
36OpenAI logoGPT-4.1 NanoOpenAIMultimodal1.0M$0.10$0.4035.2%14—
37Alibaba Qwen logoQwen3.5-FlashOSSAlibaba QwenMultimodal1.0M$0.07$0.2635.0%8—
38Meta logoLlama 4 MaverickOSSMetaMultimodal1.0M$0.20$0.7028.0%17—
39Meta logoLlama 4 ScoutOSSMetaMultimodal1.3M$0.10$0.3018.9%11—
40xAI logoGrok 4.20 BetaxAIMultimodal2.0M$2.00$6.00—0—
41xAI logoGrok 4.20 Multi-Agent BetaxAIMultimodal2.0M$2.00$6.00—0—
42Google DeepMind logoGemini 3.8 Flash (batch)Google DeepMindMultimodal1.0M$0.38$1.88—0—
43openrouter logoFusionopenrouterLLM1.0M$-1000000.00$-1000000.00—0—
44Alibaba Qwen logoQwen3.6 Plus Preview (free)OSSAlibaba QwenLLM1.0MFreeFree—0—
45
~
Claude Fable Latest
~anthropicMultimodal1.0M$10.00$50.00—0—
46NVIDIA logoNemotron 3 Ultra (free)OSSNVIDIALLM1.0MFreeFree—0—
47Alibaba Qwen logoQwen3.7 PlusOSSAlibaba QwenMultimodal1.0M$0.32$1.28—3—
48OpenAI logoGPT-5.6 Terra Pro (batch)OpenAIMultimodal1.1M$1.00$6.00—0—
49z-ai logoGLM 5.3 (batch)z-aiLLM1.0M$0.70$2.20—0—
50meituan logoLongCat 2.0meituanLLM1.0M$0.30$1.20—0—
51NVIDIA logoNemotron 3.5 Lightning (free)NVIDIALLM1.0MFreeFree—0—
52Anthropic logoClaude Opus 5AnthropicMultimodal1.0M$5.00$25.00—0—
53OpenAI logoGPT-5.6 LunaOpenAIMultimodal1.1M$0.20$1.20—0—
54Alibaba Qwen logoQwen3.8 MaxAlibaba QwenMultimodal1.0M$2.00$6.00—0—
55Anthropic logoClaude Opus 4.8 (Fast)AnthropicMultimodal1.0M$10.00$50.00—3—
56Alibaba Qwen logoQwen3.8 27BAlibaba QwenMultimodal1.0M$0.42$3.00—0—
57Alibaba Qwen logoQwen3.6 Plus (free)Alibaba QwenMultimodal1.0MFreeFree—0—
58OpenAI logoGPT-5.6 Terra ProOpenAIMultimodal1.1M$2.00$12.00—0—
59Google DeepMind logoGemini 3.5 Flash Lite (batch)Google DeepMindMultimodal1.0M$0.15$1.25—0—
60Anthropic logoClaude Opus 4.7 (Fast)AnthropicMultimodal1.0M$30.00$150.00—3—
61OpenAI logoGPT-5.6 TerraOpenAIMultimodal1.1M$2.00$12.00—0—
62Google DeepMind logoGemini 3.1 Flash LiteGoogle DeepMindMultimodal1.0M$0.25$1.50—3—
63Anthropic logoClaude Opus 5 (batch)AnthropicMultimodal1.0M$2.50$12.50—0—
64OpenAI logoGPT-5.6 Luna (batch)OpenAIMultimodal1.1M$0.10$0.60—0—
65Google DeepMind logoGemini 3.6 FlashGoogle DeepMindMultimodal1.0M$0.75$3.75—0—
66thedrummer logoUnslopNemo 12BOSSthedrummerLLM1.0M$0.40$0.40—0—
67
~
Google Gemini Pro Latest
~googleMultimodal1.0M$2.00$12.00—0—
68
~
Google Gemini Flash Latest
~googleMultimodal1.0M$1.50$9.00—0—
69
~
Anthropic Claude Sonnet Latest
~anthropicMultimodal1.0M$3.00$15.00—0—
70Google DeepMind logoGemini 3.6 Flash (batch)Google DeepMindMultimodal1.0M$0.38$1.88—0—
71
~
OpenAI GPT Latest
~openaiMultimodal1.1M$5.00$30.00—0—
72Google DeepMind logoGemma 4 26B A4B OSSGoogle DeepMindMultimodal1.0M$0.04$0.22—0—
73openrouter logoAuto Router (Beta)openrouterImage Generation2.0M$-1000000.00$-1000000.00—0—
74openrouter logoOwl AlphaopenrouterLLM1.0MFreeFree—0—
75Alibaba Qwen logoQwen Plus 0728 (thinking)OSSAlibaba QwenLLM1.0M$0.26$0.78—0—
76OpenAI logoGPT-5.6 Luna ProOpenAIMultimodal1.1M$0.20$1.20—0—
77Alibaba Qwen logoQwen3.8 FlashAlibaba QwenMultimodal1.0M$0.15$0.47—0—
78OpenAI logoGPT-5.6 Luna Pro (batch)OpenAIMultimodal1.1M$0.10$0.60—0—
79OpenAI logoGPT-5.6 SolOpenAIMultimodal1.1M$2.00$10.00—0—
80OpenAI logoGPT-5.6 Sol ProOpenAIMultimodal1.1M$2.00$10.00—0—
81
~
Claude Opus Latest
~anthropicMultimodal1.0M$5.00$25.00—0—
82OpenAI logoGPT-5.6 Terra (batch)OpenAIMultimodal1.1M$1.00$6.00—0—
83xAI logoGrok 4.20 Multi-AgentxAIMultimodal2.0M$1.25$2.50—0—
84OpenAI logoGPT-5.6 Sol Pro (batch)OpenAIMultimodal1.1M$1.00$5.00—0—
85Anthropic logoClaude Sonnet 5AnthropicMultimodal1.0M$2.00$10.00—0—
86Anthropic logoClaude Sonnet 5 (batch)AnthropicMultimodal1.0M$1.00$5.00—0—
87OpenAI logoGPT-5.6 Sol (batch)OpenAIMultimodal1.1M$1.00$5.00—0—
88Amazon logoNova 2 LiteAmazonMultimodal1.0M$0.30$2.50—1—
89z-ai logoGLM 5.2 (batch)z-aiLLM1.0M$0.70$2.20—0—
90OpenAI logoGPT-5.5 (batch)OpenAIMultimodal1.1M$2.50$15.00—0—
91Anthropic logoClaude Fable 5 (batch)AnthropicMultimodal1.0M$5.00$25.00—0—
92Anthropic logoClaude Fable 5.1 (batch)AnthropicMultimodal1.0M$5.00$25.00—0—
93Google DeepMind logoGemini 3.8 FlashGoogle DeepMindMultimodal1.0M$0.75$3.75—0—
94DeepSeek logoDeepSeek V4 Flash Vision Exp (batch)DeepSeekMultimodal1.0M$0.11$0.33—0—
95Anthropic logoClaude Opus 4.7 (batch)AnthropicMultimodal1.0M$2.50$12.50—0—
96DeepSeek logoDeepSeek V4 Flash Vision ExpDeepSeekMultimodal1.0M$0.22$0.66—0—
97Alibaba Qwen logoQwen3 Coder PlusOSSAlibaba QwenLLM1.0M$0.65$3.25—0—
98Google DeepMind logoLyria 3 Pro PreviewGoogle DeepMindLLM1.0MFreeFree—0—
99z-ai logoGLM 5.3 Flashz-aiMultimodal1.3M$0.15$0.50—0—
100minimax logoMiniMax-01OSSminimaxMultimodal1.0M$0.20$1.10—0—

Showing 1-100 of 159 · page 1/2

Top 5 by category

Specialist leaders across every modality we track

LLMs

See all →
#1OpenAI logoGPT-5.5 Pro87.8%#2OpenAI logoGPT-5.585.0%#3Anthropic logoClaude Mythos Preview81.8%#4DeepSeek logoDeepSeek V3.2 Speciale78.2%#5Anthropic logoClaude Instant78.0%

Multimodal

See all →
#1OpenAI logoGPT-5 Chat81.9%#2Alibaba Qwen logoQwen3.5 397B A17B78.4%#3Google DeepMind logoGemini 2.5 Pro Preview 05-0676.9%#4Anthropic logoClaude Fable 574.4%#5OpenAI logoo4 Mini High72.0%

Decision shortcuts

Jump from the leaderboard into comparison, provider, pricing, and benchmark paths

GPT-5.5 Pro vs Qwen3.5
Frontier comparison
Qwen3.5 vs DeepSeek V3.2
Open model comparison
OpenAI model lineup
Provider profile
Anthropic Claude lineup
Provider profile
Coding model pricing
Cost by use case
Reasoning model pricing
Cost by use case
SWE-bench Verified results
Coding benchmark
GPQA Diamond results
Reasoning benchmark

Frequently asked

Quick answers, sourced from our data

How many AI models does BenchGecko track?

BenchGecko currently tracks 1,201 AI models across 273 providers, each scored against up to 130 benchmarks. New models are added continuously and the full dataset refreshes daily.

What is the best AI model right now?

"Best" depends on the task. For general reasoning we rank by average score across 3+ benchmarks; for coding we surface ELO and SWE-bench specifically; for cost/performance we expose a "cheapest capable" metric. Use the filter bar and column sort to define your own winner, or pick a category from the mini tables below.

How is the average score calculated?

Average score is the arithmetic mean of a model's normalized benchmark scores, computed only when the model has at least one public benchmark result. Models with fewer than 3 benchmarks are excluded from the podium to avoid single-score outliers.

Where does BenchGecko get this data?

Model metadata and pricing come from OpenRouter's public API. Benchmarks are pulled from Epoch AI (CC-BY) and SWE-bench's public leaderboards. ELO ratings come from LMArena. Everything is re-normalized and cross-linked daily. See the methodology page for full provenance.

Can I use this data in my article or product?

Yes. All BenchGecko data is licensed CC BY 4.0 — attribution required. Use the "Cite this page" button above for ready-made APA, MLA, AP Style, BibTeX, and HTML embed snippets. The free API tier requires a backlink to benchgecko.ai.

How often does the data refresh?

Core model, pricing and benchmark data refreshes every 24 hours. Live status and pricing alerts can fire more frequently when upstream sources change. The "Live" pill on this page lights up when the last refresh was less than an hour ago.

Which models are open source?

Toggle the "Open Source" filter to see only models with OSS weights available. We currently count permissively — weights-available licenses like Llama, Qwen, DeepSeek, and Mistral Open count as OSS for this filter even when the weights come with usage restrictions.

How do I compare two models side by side?

Check the boxes next to any 2-4 models in the leaderboard above, then click "Compare →" in the pink action bar. You can also navigate directly to /compare/[modelA]-vs-[modelB] for shareable comparisons.

See also

Related model, benchmark, provider, and pricing indexes

Compare models
Side-by-side with charts
Benchmarks
Every eval we track
Providers
Where to run models
Pricing
Every model, every provider
AI agents
SWE-bench leaderboard
MCP servers
Tools for LLMs
AI economy
Valuations & funding
Methodology
How we score