modelgrep

Best Anthropic Models for Agents

Quick answer · Updated June 2026

Claude Fable 5 is the best Anthropic model for agents, scoring 80.6 on the Artificial Analysis Agentic Index for tool use and multi-step task completion. Claude Opus 4.8 (77.8) and Claude Opus 4.7 (71.3) round out the top three.

80.6Agentic
64.9Intelligence
$10.00Input /M
1MContext

AI models ranked by the Artificial Analysis Agentic Index — measuring multi-step tool use, planning and task completion (including Tau²-Bench). The best models for building autonomous agents and agentic workflows.

  1. 1A
    claude-fable-5
    ReasoningToolsJSON+164.9 intel · $10.00/M · 1M ctx
    80.6
    Agentic
  2. 2A
    claude-opus-4.8
    ReasoningToolsJSON+161.4 intel · $5.00/M · 59 t/s
    77.8
    Agentic
  3. 3A
    claude-opus-4.7
    ReasoningToolsJSON+157.3 intel · $5.00/M · 63 t/s
    71.3
    Agentic
  4. 4A
    claude-opus-4.6
    ReasoningToolsJSON+152.9 intel · $5.00/M · 41 t/s
    67.6
    Agentic
  5. 5A
    claude-opus-4.5
    ReasoningToolsJSON+143.1 intel · $5.00/M · 60 t/s
    59.2
    Agentic
  6. 6A
    claude-sonnet-4.6
    ReasoningToolsJSON+142.6 intel · $3.00/M · 47 t/s
    57.5
    Agentic
  7. 7A
    claude-sonnet-4.5
    ReasoningToolsJSON+137.1 intel · $3.00/M · 46 t/s
    50.6
    Agentic
  8. 8A
    claude-sonnet-4
    ReasoningToolsVision33.0 intel · $3.00/M · 49 t/s
    39.2
    Agentic
  9. 9A
    claude-haiku-4.5
    ReasoningToolsJSON+131.0 intel · $1.00/M · 82 t/s
    32.6
    Agentic
  10. 10A
    claude-3.5-haiku
    ToolsVision18.7 intel · $0.800/M · 35 t/s
    17.7
    Agentic
  11. 11A
    claude-3-haiku
    ToolsVision12.3 intel · $0.250/M · 68 t/s
    7.0
    Agentic

Frequently asked

What is the best Anthropic model for agents?

Claude Fable 5 is the best Anthropic model for agents, scoring 80.6 on the Artificial Analysis Agentic Index for tool use and multi-step task completion. Claude Opus 4.8 (77.8) and Claude Opus 4.7 (71.3) round out the top three.

What's a good alternative to Claude Fable 5?

Claude Opus 4.8 (77.8) is the closest alternative on this metric, followed by Claude Opus 4.7 (71.3). See the full ranking above for the tradeoffs.

How many Anthropic models are there?

modelgrep tracks 16 Anthropic models with live benchmarks, speed, latency and per-provider pricing, led on intelligence by Claude Fable 5. 11 of them qualify for this ranking.

More Anthropic rankings

All rankings