Pick your AI stackTry free · no accountLive · checked Nov 12, 2026

Model leaderboard

Know which AI model is best right now, what each one is best at, and what it costs.

Today's real data, every row sourced
5 free questions in the box below
Signed in? Open it in your account →

The model leaderboard today

123 models · checked Nov 12, 2026

Leaders by model type

Text
Gemini 4 Carbon
LMArena text · 2026-11-09
Decision-making
Gemini 4 Carbon
GPQA Diamond · 2026-11-02
Voice
Aurora
Text to speech
Music
Riffusion FUZZ v1.0
Music with vocals · 2026-07-31
Images
GPT-Image-2.5 Sunburst
Images from a prompt · 2026-11-09
Video
Gemini Omni 1.2 Flash
Video from a prompt · 2026-11-09

1Claude Fable 5.22,275 · $8 in, $40 out
2Gemini 4 Carbon2,266 · $2 in, $10 out
3Claude Opus 5.52,045 · $4 in, $20 out
4Gemini 4 Argon2,039 · $2 in, $10 out
5GPT-6 Astra1,927 · $10 in, $50 out
6Claude Fable 5.11,922 · $10 in, $50 out
7Claude Sonnet 5.51,887 · $2 in, $10 out
8GPT-6.1 Sol1,852 · $2 in, $10 out
9Claude Opus 51,844 · $5 in, $25 out
10Muse Wonder1,824 · $4 in, $20 out

Real workGemini 4 Carbon
Agents and toolsClaude Fable 5.2
CodingClaude Fable 5.2
Reasoning and knowledgeGemini 4 Carbon
MathClaude Fable 5.2
VisionKimi K3.1 Pro

The most for the money: nothing cheaper rates higher

Claude Fable 5.2#1 · $16 per million tokens
Gemini 4 Carbon#2 · $4 per million tokens
GPT-6.1 Terra#11 · $2 per million tokens
Gemini 3.7 Flash#17 · $1.50 per million tokens

What jumped with the newest releases

Claude Fable 5.22026-11-09
ARC-AGI-3▲ 2.6×30.2% → 79.3% vs Claude Opus 5Remote Labor Index▲ 2.3×17.9% → 41.3% vs Claude Fable 5.1SWE-Bench Pro▲ +13 pts80% → 93% vs Claude Fable 5
Doubao Seed Evolving2026-11-02
GPQA Diamond▲ +3.5 pts89.6% → 93.1% vs Doubao Seed 2.1 Turbo
Gemini 4 Carbon2026-10-28
FrontierMath Tier 4▲ +17.1 pts78.0% → 95.1% vs Gemini 4 ArgonLMArena WebDev▲ +861,682 → 1,768 vs Gemini 4 ArgonFrontierMath▲ +4.6 pts86.0% → 90.5% vs Gemini 4 Argon
GPT-6.1 Terra2026-10-26
LMArena WebDev▲ +1591,520 → 1,679 vs GPT-5.6 TerraSimpleBench▲ +23.3 pts48.9% → 72.2% vs GPT-5.6 TerraFrontierMath Tier 4▲ +21.9 pts70.7% → 92.7% vs GPT-5.6 Terra

Create a free account to open the full leaderboard

See the full rankings for text, decision-making, voice, music, images and video, with what each model costs and open-weights models ranked on their own.

Free on Explorer. No card.

Keep exploring

Answers and projections are analysis, not investment advice.

Try