Skip to content
Other

MedCoast PRM: Medical reasoning evaluation model

1 min readupdated

Built on answerdotai/ModernBERT-base, this model evaluates intermediate reasoning chains in medical question-answering (medical QA). The checkpoints are designed exclusively for inference and evaluate pairwise reasoning quality. Input is limited to 2048 tokens. Checkpoints consist of dictionaries containing fp32 model_state and configuration.

Key features

  • Based on answerdotai/ModernBERT-base
  • Evaluates intermediate reasoning chains for medical QA
  • Multiple checkpoint variants provided with different dsim and a parameters
  • Designed for inference

What you can do

  • Apply it to assess the quality of medical reasoning in QA tasks
  • Use the PRMScorer class to load and initialize checkpoints
  • Compare the accuracy of different checkpoint versions on validation data

Example

from huggingface_hub import hf_hub_download
from model.prm_scorer import PRMScorer

ckpt = hf_hub_download("zhongys97/MedCoast", "medcoast_prm_dsim0.75_a0.0.pt")
scorer = PRMScorer(api_keys_json_path="api_keys.json",
                   base_model_name="answerdotai/ModernBERT-base",
                   ft_model_path=ckpt)
Read moreDetails
Maximum input length: 2048 tokens
Best validation accuracy (val pairwise acc): 0.6093 (dsim=0.85, a=0.0)
Checkpoint format: fp32 model_state and config
Base model: answerdotai/ModernBERT-base
Discuss Telegram 1

Will it run?

Runs in BF16

Fits without quantisation.

  • BF16weights 0.6 GB · ready files1.1 GB
  • FP8 / INT8recommendedweights 0.3 GB0.8 GB
  • 4-bitweights 0.2 GB0.7 GB

Tip: go 8-bit.FP8 / INT8 is within about a percent of full precision in quality, takes half the memory and runs faster on RTX 40/50 and H100. Full precision only pays off for fine-tuning.

What to get for this model

Massed ComputeOpen Massed Compute

Your own GPU: FP8 / INT8 — from RTX 2060 · BF16 — from RTX 2060

Massed Compute hourly prices, November 2026. The discounted price applies with the coupon at checkout.

Estimate: weights plus working memory. Long videos, high resolutions and long contexts need more.

Run it on Massed Compute

The script downloads the model to the rented machine. How to run it is in the model card.

  1. 1. Rent a machine with a fitting GPU

    Open Massed Compute
  2. 2. Paste the script into the machine's terminal

    Connect over SSH (the Massed Compute panel shows the command) or open a terminal on the machine's desktop.

    HF_TOKEN / CIVITAI_TOKENThe script starts with token lines: paste yours between the quotes. The Hugging Face token is for gated models, the Civitai one for Civitai models.

    # zhongys97/MedCoast
    mkdir -p ~/models
    curl -LsSf https://astral.sh/uv/install.sh | sh
    source "$HOME/.local/bin/env"
    uv tool install "huggingface_hub[cli]"
    hf download zhongys97/MedCoast --local-dir ~/models/MedCoast
    # The model is in ~/models. How to run it is in the model card (link above).
    Add more LoRAs to MedCoast
  3. 3. Use it

    The model is in ~/models. How to run it is in the model card (link above).

Stop the machine in the panel when you are done: it bills by the minute while it runs.

How to speed it up

Picked for this model type. Choose your hardware above to see what fits you exactly.

  • FlashAttention 44.0.0b40from November 18, 2026new release

    fastest exact attention on B200

    Kernels for Blackwell (beta). Exact attention without quantisation — fine for everything, LLMs included.

    How to enable: pip install flash-attn-4; vLLM and SGLang pick it up themselves.

    Earlier generations (2)
    • FlashAttention 3 — The Hopper (H100/H200) version with FP8.
    • FlashAttention 2 — The standard for RTX 30/40/50 and A100: exact attention without the memory cost of long contexts.
  • torch.compile2.15.0from October 28, 2026new release

    10–30% for free

    Compiles the model for your GPU. Slower first run, faster afterwards.

    How to enable: model = torch.compile(model); in ComfyUI the TorchCompileModel node.

Model family

New posts about versions of ModernBERT-base (quants, finetunes, LoRAs) will show up under the bell.

Quantised versions 11All on Hugging Face
Fine-tunes 12All on Hugging Face
LoRAs and adapters 4All on Hugging Face

About the model

zhongys97/MedCoast
Author
zhongys97
Category
Other
Released
Posted

On NeyroLearn

Hugging Face

01

Ask AI about this model

Answered by Gemini 4 · the chat is kept only in this tab

Comments

Other

OmniSRE

OmniSRE