GPT-5.5 Instant on HealthBench Professional
GPT-5.5 Instant scores 0.384 on HealthBench Professional, rank 41 of 43 models on the board. The fast variant that became ChatGPT's default model in May 2026. 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.
Where it sits on the board
The models ranked around GPT-5.5 Instant, by length-adjusted score. Differences are against GPT-5.5 Instant.
| Rank | Model | Length-adjusted | Before the adjustment | Difference |
|---|---|---|---|---|
| 37 | 0.459 | 0.500 | +0.075 | |
| 38 | 0.442 | – | +0.058 | |
| 39 | 0.441 | 0.468 | +0.057 | |
| 40 | 0.396 | 0.480 | +0.012 | |
| 41 | 0.384 | 0.407 | – | |
| 42 | 0.350 | – | -0.034 | |
| 43 | 0.322 | – | -0.062 |
Result and API facts
| Rank | 41 of 43 |
|---|---|
| Length-adjusted score | 0.384 |
| Score before the length adjustment | 0.407 |
| Lab | OpenAI |
| Context window | 400K tokens |
| API price per 1M tokens | $5.00 in / $30.00 out |
| License | proprietary |
| Source | GPT-5.5 Instant System Card (system card) |
| Released | 2026-05-05 |
Source of this score
Read from GPT-5.5 Instant System Card (system card, OpenAI, 2026-05-05). Vendor-reported. Confidence: verified. Configuration: length-adjusted (40.7 unadjusted, 2,775 mean response chars); GPT-5.5 Instant system card Table 5, column GPT-5.5 INSTANT..
HealthBench Professional 37.6 (40.4, 2,973) 35.7 (38.3, 2,872) 32.9 (33.8, 2,285) 38.4 (40.7, 2,775)
Position in the field
The gap to the leader, GPT-6 Astra (Anthropic run) at 0.703, is 0.319. Directly above sits GPT-5.1 at 0.396. Directly below sits MAI-Thinking-1 at 0.350. 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-5.5 Instant score on HealthBench Professional?
GPT-5.5 Instant scores 0.384 on HealthBench Professional, which places it at rank 41 of 43 models on the board as of November 2, 2026. The number was read from GPT-5.5 Instant System Card, listed on the sources page.
How much does GPT-5.5 Instant cost to run?
GPT-5.5 Instant is priced at $5.00 per million input tokens and $30.00 per million output tokens through OpenAI's API.
Head to head
Score-difference matrixPairings with a dedicated comparison page are linked; every other difference is in the score-difference matrix.
- GPT-5.5 Instant vs GPT-6 Astra (Anthropic run)
- GPT-5.5 Instant vs Claude Sonnet 5.5
- GPT-5.5 Instant vs Claude Fable 5
- GPT-5.5 Instant vs Claude Opus 5.5
- GPT-5.5 Instant vs GPT-6 Luna (Anthropic run)
- GPT-5.5 Instant vs Claude Haiku 5.5
- GPT-5.5 Instant vs GPT-6 Astra
- GPT-5.5 Instant vs GPT-6.1 Sol
- GPT-5.5 Instant vs Claude Fable 5 (September card)
- GPT-5.5 Instant vs GPT-6.1 Terra
- GPT-5.5 Instant vs Muse Wonder Full
- GPT-5.5 Instant vs Claude Fable 5.1
- GPT-5.5 Instant vs GPT-6 Luna
- GPT-5.5 Instant vs GPT-6 Sol
- GPT-5.5 Instant vs GPT-5.6 Sol
- GPT-5.5 Instant vs Muse Wonder
- GPT-5.5 Instant vs Claude Opus 5
- GPT-5.5 Instant vs Muse Spark 1.1
- GPT-5.5 Instant vs Claude Sonnet 5
- GPT-5.5 Instant vs GPT-5.6 Terra
- GPT-5.5 Instant vs Claude Opus 4.8 (Opus 4.8 grader)
- GPT-5.5 Instant vs Grok 4.7
- GPT-5.5 Instant vs Claude Opus 4.8
- GPT-5.5 Instant vs Gemini 4 Argon
- GPT-5.5 Instant vs Muse Spark
- GPT-5.5 Instant vs GPT-5.6 Sol (August)
- GPT-5.5 Instant vs Kimi K3
- GPT-5.5 Instant vs GLM 5.3
- GPT-5.5 Instant vs Claude Opus 4.7
- GPT-5.5 Instant vs GPT-5.5
- GPT-5.5 Instant vs Gemini 3.8 Flash
- GPT-5.5 Instant vs DeepSeek V4.1 Flash
- GPT-5.5 Instant vs Grok 4.6
- GPT-5.5 Instant vs GPT-5.4
- GPT-5.5 Instant vs GPT-5
- GPT-5.5 Instant vs GPT-5.2
- GPT-5.5 Instant vs Claude Sonnet 4.6
- GPT-5.5 Instant vs GPT-5.6 Luna (August)
- GPT-5.5 Instant vs GPT-5.1
- GPT-5.5 Instant vs MAI-Thinking-1
- GPT-5.5 Instant vs Claude Haiku 4.5
Where the scores come from and how they are read is on the methodology page. The full ranking is on the leaderboard.