OverviewAll LeaderboardsMethodologyComing soon
New Chat

Work Index

Which model is the best at knowledge work?

Nov 13, 2026
1,746,812 sessions
65 models

Work Index

Frontier
AnthropicClaude Fable 5.2 (Max)1,452 on the index. 137 points above rank 2.
Daily Driver
GPT-6.1 Terra (Max)Near-frontier performance at 1/10th the price.
Efficient
GPT-6 Luna (Max)Solid performance at 1/100th the price.
1,0001,1001,2001,3001,4001,500
  1. 1,452
    Anthropic
    Claude Fable 5.2Max
  2. 1,315
    Anthropic
    Claude Opus 5.5Max
  3. 1,290
    Gemini 4 CarbonHighPreliminary
  4. 1,282
    Anthropic
    Claude Fable 5.1Max
  5. 1,272
    Anthropic
    Claude Sonnet 5.5Max
  6. 1,262
    Anthropic
    Claude Opus 5.5High
  7. 1,258
    Gemini 4 HydrogenHighPreliminary
  8. 1,254
    GPT-6 AstraMax
  9. 1,245
    GPT-6.1 SolMax
  10. 1,239
    GPT-6.1 TerraMax
  11. 1,210
    Gemini 4 ArgonHigh
  12. 1,196
    GPT-6.1 SolMedium
  13. 1,195
    Qwen4-MaxxHighPreliminary
  14. 1,194
    Meta
    Muse Wonder FullMax
  15. 1,190
    Anthropic
    Claude Sonnet 5.5High
Model RankingIndexCost per task (P50)Time per task (P50)
  1. 1AnthropicClaude Fable 5.2 (Max)1,452$11.3625.1 min
  2. 2AnthropicClaude Opus 5.5 (Max)1,315$7.8522.2 min
  3. 3Gemini 4 Carbon (High)Preliminary1,290$1.6011.4 min
  4. 4AnthropicClaude Fable 5.1 (Max)1,282$12.2520.1 min
  5. 5AnthropicClaude Sonnet 5.5 (Max)1,272$4.2319.7 min
  6. 6AnthropicClaude Opus 5.5 (High)1,262$2.646.2 min
  7. 7Gemini 4 Hydrogen (High)Preliminary1,258$1.816.1 min
  8. 8GPT-6 Astra (Max)1,254$7.4614.5 min
  9. 9GPT-6.1 Sol (Max)1,245$1.4214.5 min
  10. 10GPT-6.1 Terra (Max)1,239$1.1114.2 min
  11. 11Gemini 4 Argon (High)1,210$0.839.3 min
  12. 12GPT-6.1 Sol (Medium)1,196$0.493.9 min
  13. 13Qwen4-Max (xHigh)Preliminary1,195$2.1322.3 min
  14. 14MetaMuse Wonder Full (Max)1,194N/AN/A
  15. 15AnthropicClaude Sonnet 5.5 (High)1,190$0.993.3 min

Pareto frontier

Best index for the cost
8009001,0001,1001,2001,3001,400
MetaAnthropicAnthropic
$0.01$0.10$1$10

Median cost per task (last 14 days)

Frontier
AnthropicClaude Fable 5.2 (Max)1,452 on the index. 137 points above rank 2.
Daily Driver
GPT-6.1 Terra (Max)Near-frontier performance at 1/10th the price.
Efficient
GPT-6 Luna (Max)Solid performance at 1/100th the price.
On the frontierIndex$/task
  1. AnthropicClaude Fable 5.2 (Max)1,452$11.36
  2. AnthropicClaude Opus 5.5 (Max)1,315$7.85
  3. Gemini 4 Carbon (High)1,290$1.60
  4. GPT-6.1 Sol (Max)1,245$1.42
  5. GPT-6.1 Terra (Max)1,239$1.11
  6. Gemini 4 Argon (High)1,210$0.83
  7. GPT-6.1 Sol (Medium)1,196$0.49
  8. GPT-6.1 Terra (Medium)1,157$0.16
  9. MiMo-V2.6-Pro1,119$0.16
  10. DeepSeek-V4.1-Flash-20261017 (Max)1,116$0.14
  11. GPT-6 Luna (Max)1,107$0.08
  12. MetaMuse Ember (Max)1,104$0.02
  13. GPT-6 Luna (Low)1,011$0.01
5 models not plotted

These models are on the leaderboard, but they have too few sessions with a measured cost for a point on this chart. The numbers beside their names show their ranks.

  • 14Muse Wonder Full (Max)
  • 49Beam 1.0 501B A23B
  • 57Muse Spark 1.1
  • 60Mistral Medium 3.5
  • 61NVIDIA-Nemotron-3.5-Super-VL-120B-A12B (NVFP4)

Signal Performance

Ranking by Task CompletionScore
  1. 1AnthropicClaude Sonnet 5.5 (Max)1,307
  2. 2AnthropicClaude Opus 5.5 (Max)1,277
  3. 3Gemini 4 Carbon (High)1,261
  4. 4AnthropicClaude Fable 5.1 (Max)1,245
  5. 5Gemini 4 Hydrogen (High)1,233
  6. 6AnthropicClaude Fable 5.2 (Max)1,233
  7. 7GPT-6 Astra (Max)1,228
  8. 8Gemini 4 Argon (High)1,215
  9. 9GPT-6.1 Sol (Max)1,209
  10. 10MetaMuse Wonder Full (Max)1,203
  11. 11Qwen4-Max (xHigh)1,197
  12. 12AnthropicClaude Opus 5 (Max)1,195
  13. 13AnthropicClaude Opus 5.5 (High)1,194
  14. 14GPT-6.1 Terra (Max)1,189
  15. 15AnthropicClaude Sonnet 5.5 (High)1,169

Each rank counts all 65 models · Each bar starts at 1,100 and ends at 1,400 · Each signal is on the scale of the index

Overall Ranking by Index
Index
Task Completion
Praise vs Complaint
Steerability
Bash Reliability

Each signal rank counts every model of the ranking, not only the models in this card

Previous
Alignment Index

Which agent best aligns with your intent and reports honestly?

Next
Coding Index

Which agents are best at real coding work?

Frequently asked questions

Agent Mode

Try Agent Mode

Put these models to work on your own real tasks in Agent Mode.

Get started
How the Agent Leaderboard works

How the Agent Leaderboard works

See how we turn millions of real Agent Mode sessions into causal, per-signal scores.

Read the methodology

USE CASES

  • Chat with AI
  • Build Apps & Websites
  • Write & Edit Text
  • Search the Web
  • Generate Images
  • Generate Videos
  • Chose any model
  • Compare Models Side by Side

LEADERBOARD RANKINGS

  • Overall
  • Agent
  • Alignment Index
  • Text
  • WebDev
  • Image-to-WebDev
  • Text to Image
  • Image Edit
  • Text to Video
  • Image to Video
  • Video Edit
  • Vision
  • Document
  • Search

COMPANY

  • About Us
  • How It Works
  • Blog
  • Careers
  • Leaderboard Changelog
  • Product Changelog
  • Help Center
  • FAQ

LEGAL

  • Terms
  • Privacy
  • Cookies

FOLLOW

  • X
  • LinkedIn
  • YouTube
  • Discord

© Arena Intelligence