Skip to content
Language models

qwen3-1.7b-eka-clinical-sft-v3

compact LLM for clinical tasks

2 min readupdated

A 1.72B-parameter text model fine-tuned on the medical dataset eka-clinical-sft-train. Built by saurabhb2412 on top of the previous qwen3-1.7b-eka-clinical-sft-v2 iteration from the Qwen3 family, it is tailored for generating responses in the clinical domain. The model was fine-tuned using SFT (Supervised Fine-Tuning on curated dialogue and instruction examples) via the TRL library. Its small footprint makes it a solid fit for developers and researchers who need a lightweight local assistant or a base model for prototyping healthcare workflows.

Key features

  • Built on the Qwen3 architecture with 1.72 billion parameters (1.72B) in total.
  • Fine-tuned via SFT on the domain-specific eka-clinical-sft-train dataset.
  • Uses an up-to-date training and inference stack: TRL 1.18.0, Transformers 5.23.0, and PyTorch 2.15.0.
  • Marks the third release in a series of iterative fine-tunes following v1 and v2.

What you can do

  • Use it in conversational prototypes to answer medical and clinical questions.
  • Benchmark its output quality against the base qwen3-1.7b-eka-clinical-sft-v2 or the original Qwen3 1.7B on your own test suites.
  • Use it as a starting point for further local fine-tuning on specialized medical text-processing tasks.

Example

from transformers import pipeline

question = "What are the common symptoms of dehydration?"
generator = pipeline("text-generation", model="saurabhb2412/qwen3-1.7b-eka-clinical-sft-v3", device_map="auto")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
Read moreDetails
Base model: saurabhb2412/qwen3-1.7b-eka-clinical-sft-v2 (Qwen3 architecture, 1.72B parameters)
Training dataset: eka-clinical-sft-train
Method and tools: SFT with TRL 1.18.0, Transformers 5.23.0, PyTorch 2.15.0
Limitations and benchmarks: the model card does not list benchmark scores or VRAM requirements
Discuss Telegram 0

Will it run?

Working out the memory needs

The bot is still collecting this model's weight sizes; the estimate will appear here in a few minutes. Meanwhile you can rent a GPU with headroom.

Massed ComputeOpen Massed Compute

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.

    # saurabhb2412/qwen3-1.7b-eka-clinical-sft-v3
    # ===== Tokens: paste yours between the quotes =====
    export HF_TOKEN=""        # huggingface.co/settings/tokens, for gated models
    # ==========================================
    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 saurabhb2412/qwen3-1.7b-eka-clinical-sft-v3 --local-dir ~/models/qwen3-1.7b-eka-clinical-sft-v3
    # The model is in ~/models. How to run it is in the model card (link above).
  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.

  • MTP (multi-token prediction)0.34.0from November 16, 2026new release

    1.5–2× generation speed

    If the model has MTP layers (DeepSeek-V3/R1, Qwen3-Next, GLM, MiMo and others), it predicts several tokens per step itself.

    How to enable: vLLM / SGLang: --speculative-config '{"method": "mtp", "num_speculative_tokens": 1}'. Check the model card.

    Earlier generations (2)
    • EAGLE-3 — Speculative decoding with a small draft head. For models without their own MTP layers.
    • EAGLE-2 / Medusa — Earlier draft-head generations. Still work if that is all you find.
  • FP8 KV cache0.34.0from November 16, 2026new release

    2× more context in the same memory

    Keeps the attention cache in 8 bits, so long contexts fit where they did not.

    How to enable: vLLM: --kv-cache-dtype fp8; llama.cpp: -ctk q8_0 -ctv q8_0.

  • vLLM / SGLang0.34.0from November 16, 2026new release

    many times more tokens/s on a server

    Server engines: batching, FlashAttention, FP8 and speculative decoding out of the box.

    How to enable: vllm serve <model> — an OpenAI-compatible API.

    Earlier generations (1)
    • llama.cpp — GGUF models on any GPU, Mac or even CPU; some layers can stay in RAM.

Model family

New posts about versions of qwen3-1.7b-eka-clinical-sft-v3 (quants, finetunes, LoRAs) will show up under the bell.

No known relatives of this model yet. All derivatives on Hugging Face

About the model

saurabhb2412/qwen3-1.7b-eka-clinical-sft-v3
Author
saurabhb2412
Category
Language models
Released
Posted

On NeyroLearn

Hugging Face

00

Ask AI about this model

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

Comments