Model results
HealthBench Professional
Choose a shared evaluation protocol to compare model scores. Versions, sources and configurations stay separate.
Atlas snapshot captured 10/19/2026, 10:20:28 AM UTC. This is an import timestamp, not the evaluation date. Coverage may be incomplete.
Evaluation groups
Open a group to see its exact protocol and scores. Different configurations remain separate.
No model rankings are available in this snapshot. Reported results remain available below.
Other reported results
These reports have no model ranking. Open a group for its evidence and the reason positions are withheld.
- GPT-6.1 Solunversioned
HealthBench Professional · OpenAI system card vendor-reported (reviewed publisher results)
Single-model result · Positions withheld
Measurement date not reported
Scope · Unadjusted score; 0-100 scale; OpenAI vendor-reported GPT-6.1 Sol launch system card, Table 7, publi…
Unadjusted score; 0-100 scale; OpenAI vendor-reported GPT-6.1 Sol launch system card, Table 7, published 2026-09-29; result date not supplied; observed 2026-10-19; mean response length 3038 characters; research environment or API, with prompts, tools and effort potentially differing from production ChatGPT; exact HealthBench effort, prompts and tools not specified in Table 7; no Pro-specific claim; prior-model comparisons withheld because evaluated versions may differ from launch versions.
- GPT-6.1 Solunversioned
HealthBench Professional length-adjusted · OpenAI system card vendor-reported (reviewed publisher results)
Single-model result · Positions withheld
Measurement date not reported
Scope · Length-adjusted score; 0-100 scale; OpenAI vendor-reported GPT-6.1 Sol launch system card, Table 7, …
Length-adjusted score; 0-100 scale; OpenAI vendor-reported GPT-6.1 Sol launch system card, Table 7, published 2026-09-29; result date not supplied; observed 2026-10-19; mean response length 3038 characters; research environment or API, with prompts, tools and effort potentially differing from production ChatGPT; exact HealthBench effort, prompts and tools not specified in Table 7; no Pro-specific claim; prior-model comparisons withheld because evaluated versions may differ from launch versions.