Inference
Explore
Leaderboard
Benchmarks
Labs
Run locally
Search
⌘K
EN
English
Русский
Локально
中文
本地运行
12 Nov
2026
Find a model
General
Coding
Reasoning
Agentic
Multimodal
Fast
Low cost
Top general models
All models
#
Model
Index
Coverage
01
Claude Fable 5.2
New
Anthropic
85.6
90
%
02
Gemini 4 Carbon
Google
82.7
80
%
03
Claude Opus 5.5
Anthropic
78.6
95
%
04
Gemini 4 Argon
Google
78.5
80
%
05
GPT-6 Astra
OpenAI
78.0
100
%
05
GPT-6.1 Sol
OpenAI
78.0
100
%
07
Muse Wonder Full
Meta
77.3
80
%
08
Claude Sonnet 5.5
Anthropic
77.2
95
%
09
GPT-6.1 Terra
OpenAI
77.1
90
%
10
Claude Fable 5.1
Anthropic
76.2
95
%
Index and price
Index
Agentic
Coding
Reasoning
Multimodal
Knowledge
Multilingual
Instruction
Math
Output / 1M
Input / 1M
Speed
0
20
40
60
80
100
$0.01
$0.03
$0.1
$0.3
$1
$3
$10
$30
$100
$300
$1,000
INDEX
PARETO
OUTPUT PRICE PER 1M TOKENS · LOG SCALE
184 of 283 models
Download
PNG
Capabilities
Claude Fable 5.2
Anthropic · 37 benchmarks
Details
Agentic
82.2
2
/ 218
Coding
86.2
1
/ 209
Reasoning
86.5
1
/ 227
Multimodal
82.0
—
Knowledge
87.1
1
/ 268
Multilingual
—
—
Instruction
—
—
Math
79.3
1
/ 168
Context
tokens
PA
Pokee-Isaac 28B
10
M
Gemini 3 Pro
2
M
Grok 4.20
2
M
Grok 4.1 Fast (Reasoning)
2
M
Grok 4 Fast (Reasoning)
2
M
Gemini 1.5 Pro
2
M
Gemini 3 Pro Deep Think
2
M
Grok 4.20 Multi-agent
2
M
Llama 4 Scout
1.31
M
