HealthBench Professional

OpenAI logoGPT-6.1 Sol on HealthBench Professional

Rank 8 of 43, updated November 2, 2026.

GPT-6.1 Sol scores 0.642 on HealthBench Professional, rank 8 of 43 models on the board. 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.

8
rank of 43
0.642
length-adjusted score
0.672
score before the length adjustment
0.061
gap to the leader, GPT-6 Astra (Anthropic run)
1.05M
context window, tokens
$2.00 / $10.00
input / output price per 1M tokens

Where it sits on the board

The models ranked around GPT-6.1 Sol, by length-adjusted score. Differences are against GPT-6.1 Sol.

RankModelLength-adjustedBefore the adjustmentDifference
5OpenAI logoGPT-6 Luna (Anthropic run)0.655–+0.013
6Anthropic logoClaude Haiku 5.50.6480.710+0.006
7OpenAI logoGPT-6 Astra0.6470.682+0.005
8OpenAI logoGPT-6.1 Sol0.6420.672–
9Anthropic logoClaude Fable 5 (September card)0.6330.689-0.009
10OpenAI logoGPT-6.1 Terra0.6270.642-0.015
11Meta logoMuse Wonder Full0.623–-0.019

Result and API facts

Rank8 of 43
Length-adjusted score0.642
Score before the length adjustment0.672
LabOpenAI
Context window1.05M tokens
API price per 1M tokens$2.00 in / $10.00 out
Licenseproprietary
SourceAddendum to GPT-6 Astra System Card: GPT-6.1 Sol (system card)
Released2026-09-29

Source of this score

Read from Addendum to GPT-6 Astra System Card: GPT-6.1 Sol (system card, OpenAI, 2026-09-29). Vendor-reported. Confidence: verified. Configuration: OpenAI; length-adjusted; 64.2 (67.2, 3038) (adjusted, raw, mean response characters); addendum Table 7 refers to the GPT-6 Astra card for method.

Section 5.1, Table 7, HealthBench Professional length-adjusted, GPT-6.1 Sol columnfull entry on the sources page
EVALUATION | GPT-5.5 | GPT-5.6 SOL | GPT-5.6 TERRA | GPT-5.6 LUNA | GPT-6 ASTRA | GPT-6 SOL | GPT-6 LUNA | GPT-6.1 SOL HealthBench Professional length-adjusted | 51.8 (57.2, 3818) | 60.5 (64.1, 3228) | 57.7 (62.4, 3618) | 55.7 (59.8, 3389) | 64.7 (68.2, 3185) | 60.8 (59.5, 1573) | 60.8 (61.2, 2119) | 64.2 (67.2, 3038)

Position in the field

The gap to the leader, GPT-6 Astra (Anthropic run) at 0.703, is 0.061. Directly above sits GPT-6 Astra at 0.647. Directly below sits Claude Fable 5 (September card) at 0.633. 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-6.1 Sol score on HealthBench Professional?

GPT-6.1 Sol scores 0.642 on HealthBench Professional, which places it at rank 8 of 43 models on the board as of November 2, 2026. The number was read from Addendum to GPT-6 Astra System Card: GPT-6.1 Sol, listed on the sources page.

How much does GPT-6.1 Sol cost to run?

GPT-6.1 Sol is priced at $2.00 per million input tokens and $10.00 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-6.1 Sol vs GPT-6 Astra (Anthropic run)
    0.642 vs 0.703, GPT-6 Astra (Anthropic run) by 0.061
  • GPT-6.1 Sol vs Claude Sonnet 5.5
    0.642 vs 0.692, Claude Sonnet 5.5 by 0.050
  • GPT-6.1 Sol vs Claude Fable 5
    0.642 vs 0.660, Claude Fable 5 by 0.018
  • GPT-6.1 Sol vs Claude Opus 5.5
    0.642 vs 0.656, Claude Opus 5.5 by 0.014
  • GPT-6.1 Sol vs GPT-6 Luna (Anthropic run)
    0.642 vs 0.655, GPT-6 Luna (Anthropic run) by 0.013
  • GPT-6.1 Sol vs Claude Haiku 5.5
    0.642 vs 0.648, Claude Haiku 5.5 by 0.006
  • GPT-6.1 Sol vs GPT-6 Astra
    0.642 vs 0.647, GPT-6 Astra by 0.005
  • GPT-6.1 Sol vs Claude Fable 5 (September card)
    0.642 vs 0.633, GPT-6.1 Sol by 0.009
  • GPT-6.1 Sol vs GPT-6.1 Terra
    0.642 vs 0.627, GPT-6.1 Sol by 0.015
  • GPT-6.1 Sol vs Muse Wonder Full
    0.642 vs 0.623, GPT-6.1 Sol by 0.019
  • GPT-6.1 Sol vs Claude Fable 5.1
    0.642 vs 0.621, GPT-6.1 Sol by 0.021
  • GPT-6.1 Sol vs GPT-6 Luna
    0.642 vs 0.608, GPT-6.1 Sol by 0.034
  • GPT-6.1 Sol vs GPT-6 Sol
    0.642 vs 0.608, GPT-6.1 Sol by 0.034
  • GPT-6.1 Sol vs GPT-5.6 Sol
    0.642 vs 0.605, GPT-6.1 Sol by 0.037
  • GPT-6.1 Sol vs Muse Wonder
    0.642 vs 0.605, GPT-6.1 Sol by 0.037
  • GPT-6.1 Sol vs Claude Opus 5
    0.642 vs 0.598, GPT-6.1 Sol by 0.044
  • GPT-6.1 Sol vs Muse Spark 1.1
    0.642 vs 0.593, GPT-6.1 Sol by 0.049
  • GPT-6.1 Sol vs Claude Sonnet 5
    0.642 vs 0.578, GPT-6.1 Sol by 0.064
  • GPT-6.1 Sol vs GPT-5.6 Terra
    0.642 vs 0.577, GPT-6.1 Sol by 0.065
  • GPT-6.1 Sol vs Claude Opus 4.8 (Opus 4.8 grader)
    0.642 vs 0.574, GPT-6.1 Sol by 0.068
  • GPT-6.1 Sol vs Grok 4.7
    0.642 vs 0.567, GPT-6.1 Sol by 0.075
  • GPT-6.1 Sol vs Claude Opus 4.8
    0.642 vs 0.558, GPT-6.1 Sol by 0.084
  • GPT-6.1 Sol vs GPT-5.6 Luna
    0.642 vs 0.557, GPT-6.1 Sol by 0.085
  • GPT-6.1 Sol vs Gemini 4 Argon
    0.642 vs 0.551, GPT-6.1 Sol by 0.091
  • GPT-6.1 Sol vs Muse Spark
    0.642 vs 0.541, GPT-6.1 Sol by 0.101
  • GPT-6.1 Sol vs GPT-5.6 Sol (August)
    0.642 vs 0.540, GPT-6.1 Sol by 0.102
  • GPT-6.1 Sol vs Kimi K3
    0.642 vs 0.532, GPT-6.1 Sol by 0.110
  • GPT-6.1 Sol vs GLM 5.3
    0.642 vs 0.528, GPT-6.1 Sol by 0.114
  • GPT-6.1 Sol vs Claude Opus 4.7
    0.642 vs 0.519, GPT-6.1 Sol by 0.123
  • GPT-6.1 Sol vs GPT-5.5
    0.642 vs 0.518, GPT-6.1 Sol by 0.124
  • GPT-6.1 Sol vs Gemini 3.8 Flash
    0.642 vs 0.517, GPT-6.1 Sol by 0.125
  • GPT-6.1 Sol vs DeepSeek V4.1 Flash
    0.642 vs 0.507, GPT-6.1 Sol by 0.135
  • GPT-6.1 Sol vs Grok 4.6
    0.642 vs 0.485, GPT-6.1 Sol by 0.157
  • GPT-6.1 Sol vs GPT-5.4
    0.642 vs 0.481, GPT-6.1 Sol by 0.161
  • GPT-6.1 Sol vs GPT-5
    0.642 vs 0.462, GPT-6.1 Sol by 0.180
  • GPT-6.1 Sol vs GPT-5.2
    0.642 vs 0.459, GPT-6.1 Sol by 0.183
  • GPT-6.1 Sol vs Claude Sonnet 4.6
    0.642 vs 0.442, GPT-6.1 Sol by 0.200
  • GPT-6.1 Sol vs GPT-5.6 Luna (August)
    0.642 vs 0.441, GPT-6.1 Sol by 0.201
  • GPT-6.1 Sol vs GPT-5.1
    0.642 vs 0.396, GPT-6.1 Sol by 0.246
  • GPT-6.1 Sol vs GPT-5.5 Instant
    0.642 vs 0.384, GPT-6.1 Sol by 0.258
  • GPT-6.1 Sol vs MAI-Thinking-1
    0.642 vs 0.350, GPT-6.1 Sol by 0.292
  • GPT-6.1 Sol vs Claude Haiku 4.5
    0.642 vs 0.322, GPT-6.1 Sol by 0.320

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