Project AURIONAI Use-Cases Research & Innovation
LLM

Baichuan-M2-32B

Chinese medical LLM fine-tuned from Qwen2.5-32B; HealthBench 60.1 (developer-reported).

Baichuan AIBaichuan AIChina
Baichuan AI
Levelmodel
ClassLLM
Parameters33B
Contextnot in the notes
Released2 Sep 2025
Opennessopen weights permissive
Runs onnot in the notes
Statusnot in the notes

Benchmark Results

vendor-reported
HealthBenchdomain
vendor-reported
60.1
050100

Score in score. Higher is better.

Sep 2025 huggingface.co
HealthBench Consensusdomain
vendor-reported
91.5
050100

Score in score. Higher is better.

Sep 2025 huggingface.co
HealthBench Harddomain
vendor-reported
34.7
050100

Score in score. Higher is better.

Sep 2025 huggingface.co
Where this comes from
Every score names who ran it: vendor-reported, maintainer-run or independent. A rank is shown with the population it was ranked in (gap rule 3).

Lineage

finetune
Qwen2.5-32B to QwQ-32B: finetune, high confidenceQwen2.5-32B to Baichuan-M2-32B: finetune, high confidenceQwen2.5-32B to DeepSeek-R1-Distill-Qwen-32B: finetune, high confidenceQwen2.5-32BQwen2.5-32BAlibaba (Qwen), Sep 2024DeepSeek-R1-Distill-Qwen-32BDeepSeek-R1-Distill-Qwen-3…DeepSeek, Jan 2025QwQ-32BQwQ-32BAlibaba (Qwen), Mar 2025Baichuan-M2-32BBaichuan-M2-32BBaichuan AI, Sep 2025
Alibaba (Qwen)Qwen2.5-32B is the parent (finetune)high confidencemodel card
Where this comes from
Parent and child relations from model cards, Hugging Face base_model fields and the genealogy tables of the research notes. The tree shows two generations each way, 4 models. Confidence and evidence are the notes' own.

Industry Use Cases

Healthcare
Chinese medical LLMs

R&D Topics

Domain adaptation (vertical models)produced by

Identity

Developer
Baichuan AI
Family
not in the notes
Released
2 Sep 2025 sourcesnotes
Weights published
–
Record checked
10 Oct 2026
Where this comes from
Developer and dates from the research notes. Release date is the vendor's public announcement; weights date is when open weights were published.

Architecture and Size

Architecture
not in the notes
Total parameters
33B sources
Active parameters
–
Checkpoint parameters
–
Note
33B parameters (BF16) per model card
Context length
not in the notes
Input
text
Output
text
Where this comes from
Parameter counts as the developer states them. Where the checkpoint holds a different total, both are shown (register A16).

Openness and Access

Openness
open weights permissive
Licence
Apache License 2.0 sources
Access routes
open weights download
Where this comes from
The licence file governs over registry tags and aggregator labels.

Deployment

Runs on
not in the notes
Footprint
not in the notes
Runtimes
not in the notes
Where this comes from
Deployment tiers and runtimes as named in the notes. Footprint is the notes' figure, not a measurement by AURION.

Indonesia

Indonesian support
not stated
Regional languages
not in the notes
Jakarta hosting
unknown
Languages
not in the notes
Where this comes from
Indonesia fields are carried on every model; unknown is an allowed value.

Sources

  1. baichuan-inc/Baichuan-M2-32B , Hugging Face huggingface.co https://huggingface.co/baichuan-inc/Baichuan-M2-32B

To cite this page: Project AURION, AI Use-Cases Research & Innovation, "Baichuan-M2-32B", https://aurion.semai.id/models/baichuan-ai-baichuan-m2-32b/, records of October 2026.