HealthBench-Hallu

Baichuan-M3-235B 3.50%

GPT-5.2 3.80% leads since January 2026

Baichuan-M3-235B 3.50% leads since January 2026

Fading

Labs’ own numbers All independent runs Include claimed scores

Models answer HealthBench-style medical prompts. A pipeline splits each answer into medical claims and checks them against medical evidence. The score is a weighted hallucination rate.

Refuted claims weigh 1.0 and uncertain claims 0.5. Lower is better.

Made by: Baichuan AI
Released: January 2026
Kind: Hallucination rates
Level: Expert
Scored: Graded by a model
Answers: Open-ended
Tasks: Kept private

Strongest model tested: GPT-5.2 #50 on the capability index
Last in a launch post: January 2026, dropped before it saturated
Last model added: February 2026
Scores refreshed: November 12th, 2026

Paper arxiv.org

Model scored in lab cards

Launch post lists it, from 1 lab

System card reports it, from 1 lab

Baichuan-M3-235B leads with 3.50%

Best score per model as labs report them in their own cards. Labs pick their own settings, so these compare less cleanly than independent runs.

GPT-5.2 leads with 3.80% across all independent runs

Best score per model across the official leaderboard, Epoch AI’s own runs and any board Epoch AI copied. Each bar and row names its source.

Baichuan-M3-235B leads with 3.50%, claims included

Best score per model from every source, including labs’ own numbers and claims Epoch AI collected from reports and papers. Labs pick their own settings, so a claim can beat an independent run of the same model.

Bars Table

1 Baichuan-M3-235B 3.50%

Source: Labs’ model cards

Source: Official HealthBench-Hallu leaderboard

Sources: Labs’ model cards and official HealthBench-Hallu leaderboard

#	Model	Score	Setting	Date	Source
1	Baichuan-M3-235B	3.50%		Jan 2026	Lab model card

#	Model	Score	Setting	Date	Source
1	GPT-5.2	3.80%	high reasoning effort	Dec 2025	Official leaderboard
2	DeepSeek-V3.2	6.10%	thinking	Dec 2025	Official leaderboard
3	Gemini 3 Pro	7.10%		Nov 2025	Official leaderboard

#	Model	Score	Setting	Date	Source
1	Baichuan-M3-235B	3.50%		Jan 2026	Lab model card
2	GPT-5.2	3.80%	high reasoning effort	Dec 2025	Official leaderboard
3	DeepSeek-V3.2	6.10%	thinking	Dec 2025	Official leaderboard
4	Gemini 3 Pro	7.10%		Nov 2025	Official leaderboard

The benchmark's own leaderboard has 8 entries, 3 of them models we track. Switch the scores above to rank them together.

Number one over time Number one over time 2 models have held number one with claims included since its release in January 2026

Lab claims count from the day the card came out, independent runs from the model’s release.

OpenAI

Own leaderboard Not tested since Jan 2026

GPT-5.2 3.80%

Source: Official HealthBench-Hallu leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

On the benchmark’s own leaderboard Reported by the lab in its model card Best so far

Sources: Official HealthBench-Hallu leaderboard and labs’ model cards

Labs headline it less often now, last in January 2026

Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.

Baichuan

Date	Launch post	Score	As the post writes it
Jan 13, 2026	Baichuan-M3-235B	3.5	Hallucination Rate

One system card reports it

Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.

Date	Document	Score	As the document writes it	Section
Feb 6, 2026	Baichuan-M3-235B technical report	3.5	Hallucination Rate on HealthBench, lower is better (4.2.1 text). HealthBench-Hallu Refuted Rate 2.45 and Uncertain Rate 2.07 (Table 2)	Hallucination