HealthBench Professional

OpenAI logoGPT-6 Astra (Anthropic run) on HealthBench Professional

Rank 1 of 43, updated November 2, 2026.

GPT-6 Astra (Anthropic run) scores 0.703 on HealthBench Professional, rank 1 of 43 models on the board. The first GPT-6 model, shipped September 3, 2026, at twice the input price of GPT-5.6 Sol. HealthBench Professional scores models on 525 tasks drawn from real clinician conversations, graded criterion by criterion against physician-written rubrics on a 0 to 1 scale.

1
rank of 43
0.703
length-adjusted score
0.740
score before the length adjustment
0.011
ahead of Claude Sonnet 5.5 in second place
1.05M
context window, tokens
$10.00 / $50.00
input / output price per 1M tokens

Where it sits on the board

The models ranked around GPT-6 Astra (Anthropic run), by length-adjusted score. Differences are against GPT-6 Astra (Anthropic run).

RankModelLength-adjustedBefore the adjustmentDifference
1OpenAI logoGPT-6 Astra (Anthropic run)0.7030.740–
2Anthropic logoClaude Sonnet 5.50.6920.771-0.011
3Anthropic logoClaude Fable 50.6600.703-0.043
4Anthropic logoClaude Opus 5.50.6560.771-0.047
5OpenAI logoGPT-6 Luna (Anthropic run)0.655–-0.048
6Anthropic logoClaude Haiku 5.50.6480.710-0.055
7OpenAI logoGPT-6 Astra0.6470.682-0.056

Result and API facts

Rank1 of 43
Length-adjusted score0.703
Score before the length adjustment0.740
LabOpenAI
Context window1.05M tokens
API price per 1M tokens$10.00 in / $50.00 out
Licenseproprietary
SourceClaude Sonnet 5.5 System Card (system card)
Released2026-09-03

Source of this score

Read from Claude Sonnet 5.5 System Card (system card, Anthropic, 2026-09-28). Independent run. Confidence: verified. Configuration: Anthropic public-API reproduction; max effort; no system prompt; Opus 4.8 grader; length-adjusted; raw 74.0%.

p. 138, paragraph above Figure 8.15.B, HealthBench Professional after length adjustment.full entry on the sources page
On HealthBench Professional at max effort, length adjustment changes the ranking. After: GPT-6 Astra (70.3%) > Claude Sonnet 5.5 (69.2%) > Claude Opus 5.5 (65.6%) > Claude Fable 5.1 (62.1%).

Position in the field

GPT-6 Astra (Anthropic run) leads the leaderboard, 0.011 ahead of Claude Sonnet 5.5 in second place. The scores on this page are compiled from published documents rather than from one controlled run, so a small gap between 2 models can reflect a difference in grader version, reasoning effort, or deployment setting as well as a difference in capability.

What does GPT-6 Astra (Anthropic run) score on HealthBench Professional?

GPT-6 Astra (Anthropic run) scores 0.703 on HealthBench Professional, which places it at rank 1 of 43 models on the board as of November 2, 2026. The number was read from Claude Sonnet 5.5 System Card, listed on the sources page.

How much does GPT-6 Astra (Anthropic run) cost to run?

GPT-6 Astra (Anthropic run) is priced at $10.00 per million input tokens and $50.00 per million output tokens through OpenAI's API.

Pairings with a dedicated comparison page are linked; every other difference is in the score-difference matrix.

  • GPT-6 Astra (Anthropic run) vs Claude Sonnet 5.5
    0.703 vs 0.692, GPT-6 Astra (Anthropic run) by 0.011
  • GPT-6 Astra (Anthropic run) vs Claude Fable 5
    0.703 vs 0.660, GPT-6 Astra (Anthropic run) by 0.043
  • GPT-6 Astra (Anthropic run) vs Claude Opus 5.5
    0.703 vs 0.656, GPT-6 Astra (Anthropic run) by 0.047
  • GPT-6 Astra (Anthropic run) vs GPT-6 Luna (Anthropic run)
    0.703 vs 0.655, GPT-6 Astra (Anthropic run) by 0.048
  • GPT-6 Astra (Anthropic run) vs Claude Haiku 5.5
    0.703 vs 0.648, GPT-6 Astra (Anthropic run) by 0.055
  • GPT-6 Astra (Anthropic run) vs GPT-6 Astra
    0.703 vs 0.647, GPT-6 Astra (Anthropic run) by 0.056
  • GPT-6 Astra (Anthropic run) vs GPT-6.1 Sol
    0.703 vs 0.642, GPT-6 Astra (Anthropic run) by 0.061
  • GPT-6 Astra (Anthropic run) vs Claude Fable 5 (September card)
    0.703 vs 0.633, GPT-6 Astra (Anthropic run) by 0.070
  • GPT-6 Astra (Anthropic run) vs GPT-6.1 Terra
    0.703 vs 0.627, GPT-6 Astra (Anthropic run) by 0.076
  • GPT-6 Astra (Anthropic run) vs Muse Wonder Full
    0.703 vs 0.623, GPT-6 Astra (Anthropic run) by 0.080
  • GPT-6 Astra (Anthropic run) vs Claude Fable 5.1
    0.703 vs 0.621, GPT-6 Astra (Anthropic run) by 0.082
  • GPT-6 Astra (Anthropic run) vs GPT-6 Luna
    0.703 vs 0.608, GPT-6 Astra (Anthropic run) by 0.095
  • GPT-6 Astra (Anthropic run) vs GPT-6 Sol
    0.703 vs 0.608, GPT-6 Astra (Anthropic run) by 0.095
  • GPT-6 Astra (Anthropic run) vs GPT-5.6 Sol
    0.703 vs 0.605, GPT-6 Astra (Anthropic run) by 0.098
  • GPT-6 Astra (Anthropic run) vs Muse Wonder
    0.703 vs 0.605, GPT-6 Astra (Anthropic run) by 0.098
  • GPT-6 Astra (Anthropic run) vs Claude Opus 5
    0.703 vs 0.598, GPT-6 Astra (Anthropic run) by 0.105
  • GPT-6 Astra (Anthropic run) vs Muse Spark 1.1
    0.703 vs 0.593, GPT-6 Astra (Anthropic run) by 0.110
  • GPT-6 Astra (Anthropic run) vs Claude Sonnet 5
    0.703 vs 0.578, GPT-6 Astra (Anthropic run) by 0.125
  • GPT-6 Astra (Anthropic run) vs GPT-5.6 Terra
    0.703 vs 0.577, GPT-6 Astra (Anthropic run) by 0.126
  • GPT-6 Astra (Anthropic run) vs Claude Opus 4.8 (Opus 4.8 grader)
    0.703 vs 0.574, GPT-6 Astra (Anthropic run) by 0.129
  • GPT-6 Astra (Anthropic run) vs Grok 4.7
    0.703 vs 0.567, GPT-6 Astra (Anthropic run) by 0.136
  • GPT-6 Astra (Anthropic run) vs Claude Opus 4.8
    0.703 vs 0.558, GPT-6 Astra (Anthropic run) by 0.145
  • GPT-6 Astra (Anthropic run) vs GPT-5.6 Luna
    0.703 vs 0.557, GPT-6 Astra (Anthropic run) by 0.146
  • GPT-6 Astra (Anthropic run) vs Gemini 4 Argon
    0.703 vs 0.551, GPT-6 Astra (Anthropic run) by 0.152
  • GPT-6 Astra (Anthropic run) vs Muse Spark
    0.703 vs 0.541, GPT-6 Astra (Anthropic run) by 0.162
  • GPT-6 Astra (Anthropic run) vs GPT-5.6 Sol (August)
    0.703 vs 0.540, GPT-6 Astra (Anthropic run) by 0.163
  • GPT-6 Astra (Anthropic run) vs Kimi K3
    0.703 vs 0.532, GPT-6 Astra (Anthropic run) by 0.171
  • GPT-6 Astra (Anthropic run) vs GLM 5.3
    0.703 vs 0.528, GPT-6 Astra (Anthropic run) by 0.175
  • GPT-6 Astra (Anthropic run) vs Claude Opus 4.7
    0.703 vs 0.519, GPT-6 Astra (Anthropic run) by 0.184
  • GPT-6 Astra (Anthropic run) vs GPT-5.5
    0.703 vs 0.518, GPT-6 Astra (Anthropic run) by 0.185
  • GPT-6 Astra (Anthropic run) vs Gemini 3.8 Flash
    0.703 vs 0.517, GPT-6 Astra (Anthropic run) by 0.186
  • GPT-6 Astra (Anthropic run) vs DeepSeek V4.1 Flash
    0.703 vs 0.507, GPT-6 Astra (Anthropic run) by 0.196
  • GPT-6 Astra (Anthropic run) vs Grok 4.6
    0.703 vs 0.485, GPT-6 Astra (Anthropic run) by 0.218
  • GPT-6 Astra (Anthropic run) vs GPT-5.4
    0.703 vs 0.481, GPT-6 Astra (Anthropic run) by 0.222
  • GPT-6 Astra (Anthropic run) vs GPT-5
    0.703 vs 0.462, GPT-6 Astra (Anthropic run) by 0.241
  • GPT-6 Astra (Anthropic run) vs GPT-5.2
    0.703 vs 0.459, GPT-6 Astra (Anthropic run) by 0.244
  • GPT-6 Astra (Anthropic run) vs Claude Sonnet 4.6
    0.703 vs 0.442, GPT-6 Astra (Anthropic run) by 0.261
  • GPT-6 Astra (Anthropic run) vs GPT-5.6 Luna (August)
    0.703 vs 0.441, GPT-6 Astra (Anthropic run) by 0.262
  • GPT-6 Astra (Anthropic run) vs GPT-5.1
    0.703 vs 0.396, GPT-6 Astra (Anthropic run) by 0.307
  • GPT-6 Astra (Anthropic run) vs GPT-5.5 Instant
    0.703 vs 0.384, GPT-6 Astra (Anthropic run) by 0.319
  • GPT-6 Astra (Anthropic run) vs MAI-Thinking-1
    0.703 vs 0.350, GPT-6 Astra (Anthropic run) by 0.353
  • GPT-6 Astra (Anthropic run) vs Claude Haiku 4.5
    0.703 vs 0.322, GPT-6 Astra (Anthropic run) by 0.381

Where the scores come from and how they are read is on the methodology page. The full ranking is on the leaderboard.