Skip to main content
BenchLM
Data
800 families · October 2026

All 954 AI Models Compared

Browse 954 AI language models across 800 families, ranked by public leaderboard scores across agentic, coding, reasoning, knowledge, multimodal, grounded, instruction-following, and multilingual tasks. Every model and SKU has its own row, linked to the other versions in its family.

Describe the task and constraints. This selector recommends models; it cannot carry out the task or access your files and accounts.

Directory filters

Showing 25 of 224 ranked models

Claude Opus 5.5
#1Claude Opus 5.5

Anthropic

1M · Reasoning · Current

86.56

Supported

$12.00

Claude Sonnet 5.5
#5Claude Sonnet 5.5

Anthropic

1M · Reasoning · Current

82.18

Supported

$6.00

Gemini 4 Argon
#7Gemini 4 Argon

Google

Family: Gemini 4

1M · Reasoning · Current

80.97

Supported

$6.00

Claude Opus 5
#8Claude Opus 5

Anthropic

Context pending · Reasoning · Superseded

79.63

Supported

$15.00

Claude Haiku 5.5
#16Claude Haiku 5.5

Anthropic

1M · Reasoning · Current

70.04

Supported

$0.30

Claude Opus 4.8
#17Claude Opus 4.8

Anthropic

1M · Reasoning · Superseded

70.01

Supported

$15.00

Kimi K3
#18Kimi K3

Moonshot AI

1.05M · Reasoning · Current

69.87

Supported

$9.00

Step 5 Preview
#20Step 5 Preview

StepFun

Family: Step 5

1M · Reasoning · Current

69.44

$1.85

Ling 3.1 Flash
#25Ling 3.1 Flash

InclusionAI

Family: Ling 3.1

262K · Reasoning · Current

68.03

—

25 of 224 shown