Skip to content
Computer vision

DesignCoder 4B

SFT-чекпоинт для дизайн-кода

1 min read

DesignCoder 4B от chjdl1816 — это merged BF16-модель на 4,54 млрд параметров, полученная из LoRA SFT-чекпоинта Qwen3.5-4B. Это SFT-модель, а не RL-чекпоинт; AdamW-версия выбрана для инициализации DesignCoder RL.

Ключевые особенности

  • Полный экспорт: загрузка не требует PEFT или отдельной базовой модели.
  • LoRA rank 64, alpha 128, dropout 0.05; 2 эпохи, step 12588.
  • Обучение с AdamW, LR 1e-4, cosine schedule, effective batch 8.
  • Training cutoff — 16384 токена; токенизатор и chat template из чекпоинта.
  • В репозитории также лежат исходные adapter-веса в adapter/.

Пример

from transformers import AutoTokenizer, Qwen3_5ForConditionalGeneration

model_id = "chjdl1816/designcoder-mc"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = Qwen3_5ForConditionalGeneration.from_pretrained(
    model_id, dtype="auto", device_map="auto"
)
ПодробнееБенчмарки в карточке не заявлены; GPU-инференс и vLLM end-to-end RL-оценка не проводились.
Экспорт проверен с Transformers 5.8.0, PEFT 0.18.1 и PyTorch 2.6.0.
Лицензия — Apache-2.0, как у базовой Qwen3.5-4B.
Не применяйте adapter к уже merged-модели в корне репозитория.
Discuss Telegram 0 23

Will it run?

Runs in BF16

Fits without quantisation.

  • BF16weights 8.5 GB · ready files9.8 GB
  • FP8 / INT8recommendedweights 4.2 GB5.2 GB
  • 4-bitweights 2.4 GB3.1 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 3080

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.

    # chjdl1816/designcoder-mc
    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 chjdl1816/designcoder-mc --local-dir ~/models/designcoder-mc
    # The model is in ~/models. How to run it is in the model card (link above).
    Add more LoRAs to designcoder-mc
  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

  1. Qwen3.5-4B-Basefamily root
  2. Qwen3.5-4B
  3. designcoder-mc

New posts about versions of Qwen3.5-4B-Base (quants, finetunes, LoRAs) will show up under the bell.

Quantised versions 12All on Hugging Face
Fine-tunes 12All on Hugging Face
LoRAs and adapters 12All on Hugging Face
Merges 1All on Hugging Face

About the model

chjdl1816/designcoder-mc
Author
chjdl1816
Category
Computer vision
Released
Posted

On NeyroLearn

Hugging Face

230

Ask AI about this model

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

Comments