Text Generation
Transformers
Safetensors
English
Chinese
qwen3_5
image-text-to-text
qwen
lora
reinforcement-learning
fp8
agentic-coding
scientific-computing
physics
materials-science
genetics
conversational
Instructions to use HoomanBio2021/LabOpen-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use HoomanBio2021/LabOpen-v1 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="HoomanBio2021/LabOpen-v1") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] pipe(text=messages)# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("HoomanBio2021/LabOpen-v1") model = AutoModelForMultimodalLM.from_pretrained("HoomanBio2021/LabOpen-v1", device_map="auto") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] inputs = processor.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=256) print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use HoomanBio2021/LabOpen-v1 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "HoomanBio2021/LabOpen-v1" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "HoomanBio2021/LabOpen-v1", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/HoomanBio2021/LabOpen-v1
- SGLang
How to use HoomanBio2021/LabOpen-v1 with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "HoomanBio2021/LabOpen-v1" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "HoomanBio2021/LabOpen-v1", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "HoomanBio2021/LabOpen-v1" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "HoomanBio2021/LabOpen-v1", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use HoomanBio2021/LabOpen-v1 with Docker Model Runner:
docker model run hf.co/HoomanBio2021/LabOpen-v1
Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -24,7 +24,7 @@ LabOpen v1 adds six verifiable capability domains (fluid physics, materials mixt
|
|
| 24 |
|
| 25 |
## 项目定位与理念
|
| 26 |
|
| 27 |
-
Qwen3.8-27B-LabOpen
|
| 28 |
|
| 29 |
项目并不以取代通用大模型为目标,而是希望在保留通用模型综合能力的基础上,进一步增强其对科学专业知识、领域问题及研究任务的适应能力。
|
| 30 |
|
|
@@ -75,7 +75,7 @@ Use the **merged weights** for serving (vLLM/SGLang). Use the **adapter** on top
|
|
| 75 |
| DeepSeek v4.1 flash | 6/10 | 48/78 (62%) | #4 |
|
| 76 |
| Qwen3.8-27B (base) | 8/10 | 25/78 (32%) | #5 |
|
| 77 |
|
| 78 |
-
A 27B + r4 LoRA reaching frontier-model agentic-coding territory
|
| 79 |
|
| 80 |
## Usage
|
| 81 |
|
|
@@ -127,7 +127,7 @@ Without a tag the model keeps base behavior (code / function calling need no tag
|
|
| 127 |
## Technical notes
|
| 128 |
|
| 129 |
- **Architecture**: Qwen3.8-27B hybrid — 64 decoder layers: 48 linear-attention + 16 full-attention layers.
|
| 130 |
-
- **LoRA**: r=4, α=4 (scaling 1.0), dropout 0, on 496 projection modules across
|
| 131 |
- **Merge math**: per module `W = codes × scale(blockwise)` dequantized to FP32, LoRA delta `(B·A)·scaling` added, re-quantized per 128×128 block (`scale = amax/448`). Mean re-quantization relative error 0.33% — the same order as the base's own FP8 storage noise. Equivalence probes (greedy, 3 prompts × 48 tokens, adapter vs merged): see `MERGE_REPORT.md`.
|
| 132 |
- **Training substrate**: a local FP8 copy of Qwen3.8-27B (community redistribution). Bit-equivalence with `Qwen/Qwen3.8-27B-FP8` was not verified; the adapter config points to the official release for loading.
|
| 133 |
|
|
@@ -182,7 +182,7 @@ LabOpen v1 在 [Qwen/Qwen3.8-27B-FP8](https://huggingface.co/Qwen/Qwen3.8-27B-FP
|
|
| 182 |
|
| 183 |
## 项目定位与理念
|
| 184 |
|
| 185 |
-
Qwen3.8-27B-LabOpen
|
| 186 |
|
| 187 |
项目并不以取代通用大模型为目标,而是希望在保留通用模型综合能力的基础上,进一步增强其对科学专业知识、领域问题及研究任务的适应能力。
|
| 188 |
|
|
@@ -225,7 +225,7 @@ Qwen3.8-27B-LabOpen - v1 是基于 Qwen3.8-27B 开源基础模型开展领域微
|
|
| 225 |
| DeepSeek v4.1 flash | 6/10 | 48/78(62%) | #4 |
|
| 226 |
| Qwen3.8-27B(基座) | 8/10 | 25/78(32%) | #5 |
|
| 227 |
|
| 228 |
-
27B + r4 LoRA 以
|
| 229 |
|
| 230 |
## 使用方法
|
| 231 |
|
|
@@ -277,13 +277,13 @@ tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen3.8-27B-FP8")
|
|
| 277 |
## 技术说明
|
| 278 |
|
| 279 |
- **架构**:Qwen3.8-27B 混合——64 个解码层:48 层线性注意力 + 16 层全注意力。
|
| 280 |
-
- **LoRA**:r=4,α=4(缩放 1.0),dropout 0,覆盖
|
| 281 |
- **合并数学**:逐模块 `W = codes × scale`(分块)反量化至 FP32,加入 LoRA 增量 `(B·A)·scaling`,再按 128×128 块重量化(`scale = amax/448`)。平均重量化相对误差 0.33%——与基座自身 FP8 存储噪声同量级。等价探针(greedy,3 提示 × 48 token,适配器 vs 合并):见 `MERGE_REPORT.md`。
|
| 282 |
- **训练基材**:Qwen3.8-27B 的本地 FP8 副本(社区再分发)。未与 `Qwen/Qwen3.8-27B-FP8` 做逐位等价核验;适配器配置指向官方发布以便加载。
|
| 283 |
|
| 284 |
## 训练方法(摘要)
|
| 285 |
|
| 286 |
-
15 个强化窗口(G69–G84),运行于带
|
| 287 |
|
| 288 |
## 诚实边界
|
| 289 |
|
|
|
|
| 24 |
|
| 25 |
## 项目定位与理念
|
| 26 |
|
| 27 |
+
Qwen3.8-27B-LabOpen v1 是基于 Qwen3.8-27B 开源基础模型开展领域微调与优化的研究模型,重点面向编程、物理学、生命科学、材料科学等科学研究领域。
|
| 28 |
|
| 29 |
项目并不以取代通用大模型为目标,而是希望在保留通用模型综合能力的基础上,进一步增强其对科学专业知识、领域问题及研究任务的适应能力。
|
| 30 |
|
|
|
|
| 75 |
| DeepSeek v4.1 flash | 6/10 | 48/78 (62%) | #4 |
|
| 76 |
| Qwen3.8-27B (base) | 8/10 | 25/78 (32%) | #5 |
|
| 77 |
|
| 78 |
+
A 27B + r4 LoRA reaching frontier-model agentic-coding territory in 15 training windows is the headline result; see `EVALUATION.md` for gaps that remain (first-pass rate, general knowledge vs 100B+ models).
|
| 79 |
|
| 80 |
## Usage
|
| 81 |
|
|
|
|
| 127 |
## Technical notes
|
| 128 |
|
| 129 |
- **Architecture**: Qwen3.8-27B hybrid — 64 decoder layers: 48 linear-attention + 16 full-attention layers.
|
| 130 |
+
- **LoRA**: r=4, α=4 (scaling 1.0), dropout 0, on 496 projection modules across 12 module types (q/k/v/o/gate/up/down_proj, in_proj_a/b/qkv/z, out_proj) = 992 tensors, ~112 MB.
|
| 131 |
- **Merge math**: per module `W = codes × scale(blockwise)` dequantized to FP32, LoRA delta `(B·A)·scaling` added, re-quantized per 128×128 block (`scale = amax/448`). Mean re-quantization relative error 0.33% — the same order as the base's own FP8 storage noise. Equivalence probes (greedy, 3 prompts × 48 tokens, adapter vs merged): see `MERGE_REPORT.md`.
|
| 132 |
- **Training substrate**: a local FP8 copy of Qwen3.8-27B (community redistribution). Bit-equivalence with `Qwen/Qwen3.8-27B-FP8` was not verified; the adapter config points to the official release for loading.
|
| 133 |
|
|
|
|
| 182 |
|
| 183 |
## 项目定位与理念
|
| 184 |
|
| 185 |
+
Qwen3.8-27B-LabOpen v1 是基于 Qwen3.8-27B 开源基础模型开展领域微调与优化的研究模型,重点面向编程、物理学、生命科学、材料科学等科学研究领域。
|
| 186 |
|
| 187 |
项目并不以取代通用大模型为目标,而是希望在保留通用模型综合能力的基础上,进一步增强其对科学专业知识、领域问题及研究任务的适应能力。
|
| 188 |
|
|
|
|
| 225 |
| DeepSeek v4.1 flash | 6/10 | 48/78(62%) | #4 |
|
| 226 |
| Qwen3.8-27B(基座) | 8/10 | 25/78(32%) | #5 |
|
| 227 |
|
| 228 |
+
27B + r4 LoRA 以 15 个训练窗口进入前沿模型的智能体编程区间,这是本发布的头条结果;仍存在的差距(首过率、与 100B+ 模型的通用知识差距)见 `EVALUATION.md`。
|
| 229 |
|
| 230 |
## 使用方法
|
| 231 |
|
|
|
|
| 277 |
## 技术说明
|
| 278 |
|
| 279 |
- **架构**:Qwen3.8-27B 混合——64 个解码层:48 层线性注意力 + 16 层全注意力。
|
| 280 |
+
- **LoRA**:r=4,α=4(缩放 1.0),dropout 0,覆盖 12 类模块(q/k/v/o/gate/up/down_proj、in_proj_a/b/qkv/z、out_proj)共 496 个投影模块 = 992 张量,约 112 MB。
|
| 281 |
- **合并数学**:逐模块 `W = codes × scale`(分块)反量化至 FP32,加入 LoRA 增量 `(B·A)·scaling`,再按 128×128 块重量化(`scale = amax/448`)。平均重量化相对误差 0.33%——与基座自身 FP8 存储噪声同量级。等价探针(greedy,3 提示 × 48 token,适配器 vs 合并):见 `MERGE_REPORT.md`。
|
| 282 |
- **训练基材**:Qwen3.8-27B 的本地 FP8 副本(社区再分发)。未与 `Qwen/Qwen3.8-27B-FP8` 做逐位等价核验;适配器配置指向官方发布以便加载。
|
| 283 |
|
| 284 |
## 训练方法(摘要)
|
| 285 |
|
| 286 |
+
15 个强化窗口(G69–G84),运行于带精确权重/更新契约的冻结 FP8 服务栈:组相对优势 + k=8 采样 + 机械可验证奖励;零起点域的 oracle 注入;冠军检查点的验证蒸馏;冻结域标记;多步组合的子步优势;逐窗 heldout 门与保持保护。单种子单次运行——本卡中的全部能力声明均来自训练未见过的评测集。
|
| 287 |
|
| 288 |
## 诚实边界
|
| 289 |
|