ObviousBench

PUBLIC MODEL RELIABILITY · LAB VIEW

Anthropic

68 public configurations across 16 model families in the 2026-08-13 release. The global cost frontier stays available as context.

GLOBAL REFERENCE RETAINED

Where Anthropic sits in the complete public field

The lab is highlighted without changing the frontier calculation or zoom. This preserves the comparison frame that a provider-only chart would otherwise hide.

Anthropic in the global fieldAll 555 public configurations on the same global log-cost and pass3 scale as the homepage frontier. Public field AnthropicGlobal Pareto
Anthropic configurations highlighted in the global ObviousBench cost and reliability field 100% 95% 90% 80% 60% 40% 20% 10% Claude 3 Haiku · None / disabled · 44.4% · $0.008526 Claude 3.5 Haiku · None / disabled · 59% · $0.02929 Claude Haiku 4.5 · None / disabled · 65.3% · $0.04191 Claude Opus 4 · None / disabled · 90.3% · $0.7667 Claude Opus 4.1 · None / disabled · 91% · $0.7664 Claude Opus 4.5 · None / disabled · 81.9% · $0.1908 Claude Opus 4.8 · None / disabled · 84.7% · $0.2029 Claude Sonnet 4 · None / disabled · 57.6% · $0.1709 Claude Sonnet 4.5 · None / disabled · 68.1% · $0.1862 Claude Sonnet 4.6 · None / disabled · 79.9% · $0.1292 Claude Opus 4 · Reasoning · 98.6% · $4.962 Claude Opus 4.1 · Reasoning · 98.6% · $4.925 Claude Fable 5 · Low · 97.9% · $0.6193 Claude Fable 5.1 · Low · 88.9% · $0.5002 Claude Haiku 4.5 · Low · 91.7% · $0.3611 Claude Opus 4.5 · Low · 99.3% · $1.602 Claude Opus 4.6 · Low · 90.3% · $0.1779 Claude Opus 4.7 · Low · 84% · $0.235 Claude Opus 4.8 · Low · 95.1% · $0.3032 Claude Sonnet 4 · Low · 94.4% · $1.035 Claude Sonnet 4.5 · Low · 95.8% · $0.9472 Claude Sonnet 4.6 · Low · 79.2% · $0.1211 Claude Sonnet 5 · Low · 82.6% · $0.09626 Claude Fable 5 · Medium · 97.9% · $0.7192 Claude Fable 5.1 · Medium · 90.3% · $0.4691 Claude Haiku 4.5 · Medium · 92.4% · $0.3833 Claude Opus 4.5 · Medium · 99.3% · $1.678 Claude Opus 4.6 · Medium · 89.6% · $0.204 Claude Opus 4.7 · Medium · 85.4% · $0.2348 Claude Opus 4.8 · Medium · 93.1% · $0.32 Claude Sonnet 4 · Medium · 95.1% · $1.142 Claude Sonnet 4.5 · Medium · 96.5% · $0.9708 Claude Sonnet 4.6 · Medium · 88.2% · $0.189 Claude Sonnet 5 · Medium · 80.6% · $0.09943 Claude Fable 5 · High · 98.6% · $0.8537 Claude Fable 5.1 · High · 88.9% · $0.5313 Claude Haiku 4.5 · High · 90.3% · $0.4208 Claude Opus 4.5 · High · 99.3% · $1.691 Claude Opus 4.6 · High · 97.9% · $0.6526 Claude Opus 4.7 · High · 86.8% · $0.2455 Claude Opus 4.8 · High · 94.4% · $0.3542 Claude Sonnet 4 · High · 95.8% · $1.236 Claude Sonnet 4.5 · High · 97.2% · $1.089 Claude Sonnet 4.6 · High · 96.5% · $0.5734 Claude Sonnet 5 · High · 86.8% · $0.1293 Claude Fable 5 · Extra high · 98.6% · $1.016 Claude Fable 5.1 · Extra high · 96.5% · $0.7154 Claude Haiku 4.5 · Extra high · 93.1% · $0.4391 Claude Opus 4.5 · Extra high · 98.6% · $1.71 Claude Opus 4.6 · Extra high · 97.9% · $0.6578 Claude Opus 4.7 · Extra high · 91.7% · $0.2859 Claude Opus 4.8 · Extra high · 96.5% · $0.5078 Claude Sonnet 4 · Extra high · 95.8% · $1.215 Claude Sonnet 4.5 · Extra high · 97.2% · $1.074 Claude Sonnet 5 · Extra high · 90.3% · $0.1759 Claude Fable 5 · Max · 99.3% · $2.123 Claude Fable 5.1 · Max · 99.3% · $1.621 Claude Opus 4.6 · Max · 97.2% · $0.8082 Claude Opus 4.7 · Max · 86.8% · $0.3351 Claude Opus 4.8 · Max · 99.3% · $1.004 Claude Sonnet 4.6 · Max · 96.5% · $0.822 Claude Sonnet 5 · Max · 96.5% · $0.5097 Claude Opus 5 · None / disabled · 80.6% · $0.2455 Claude Opus 5 · Low · 98.6% · $0.433 Claude Opus 5 · Medium · 99.3% · $0.503 Claude Opus 5 · High · 100% · $0.5998 Claude Opus 5 · Extra high · 99.3% · $0.6583 Claude Opus 5 · Max · 99.3% · $0.7376 More expensive Cheaper

This is a fixed global reference, not a recomputed Anthropic frontier. Hover a highlighted point for its observed setting, then select it to continue in the interactive cost frontier.

CURATED DETAIL

Decision pages for Anthropic

These highlighted cards have an indexable evidence page. The rest remain in the full explorer rather than being turned into thin pages.

FULL LAB VIEW

All observed Anthropic model families

Sorted by highest observed answer pass3, then the lowest cost at that score. Release dates are not inferred because this release data does not encode a consistent vendor release-date field.

Claude Opus 5Best observed 100% · $0.5998
Claude Opus 4.8Best observed 99.3% · $1.004
Claude Opus 4.5Best observed 99.3% · $1.602
Claude Fable 5.1Best observed 99.3% · $1.621
Claude Fable 5Best observed 99.3% · $2.123
Claude Opus 4.1Best observed 98.6% · $4.925
Claude Opus 4Best observed 98.6% · $4.962
Claude Opus 4.6Best observed 97.9% · $0.6526
Claude Sonnet 4.5Best observed 97.2% · $1.074
Claude Sonnet 5Best observed 96.5% · $0.5097
Claude Sonnet 4.6Best observed 96.5% · $0.5734
Claude Sonnet 4Best observed 95.8% · $1.215
Claude Haiku 4.5Best observed 93.1% · $0.4391
Claude Opus 4.7Best observed 91.7% · $0.2859
Claude 3.5 HaikuBest observed 59% · $0.02929
Claude 3 HaikuBest observed 44.4% · $0.008526