Spaces:
Running
Running
Add NVFP4 project and GitHub links to organization card
Browse filesPresent the Qwen4-27B NVFP4 preparation toolkit, release checklist, and progress link. Clearly state that model weights and Qwen4 compatibility are not yet validated.
README.md
CHANGED
|
@@ -13,6 +13,20 @@ pinned: false
|
|
| 13 |
|
| 14 |
IteraCompute Artificial Intelligence Technology (Chongqing) Co., Ltd. provides reliable, usage-based inference for language models through OpenAI-compatible APIs. We focus on predictable integration, streaming responses, tool use, structured outputs, and privacy-conscious processing.
|
| 15 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
## Hugging Face Inference Provider
|
| 17 |
|
| 18 |
Our Hugging Face Inference Provider integration is currently being prepared for review. Initial conversational model mappings include:
|
|
@@ -27,6 +41,7 @@ The integration supports conversational inference, streaming, tool calling, stru
|
|
| 27 |
## Resources
|
| 28 |
|
| 29 |
- [Website](https://iteracompute.com/)
|
|
|
|
| 30 |
- [API documentation](https://iteracompute.com/docs.html)
|
| 31 |
- [Privacy policy](https://iteracompute.com/privacy.html)
|
| 32 |
- Provider partnerships: [partnerships@iteracompute.com](mailto:partnerships@iteracompute.com)
|
|
|
|
| 13 |
|
| 14 |
IteraCompute Artificial Intelligence Technology (Chongqing) Co., Ltd. provides reliable, usage-based inference for language models through OpenAI-compatible APIs. We focus on predictable integration, streaming responses, tool use, structured outputs, and privacy-conscious processing.
|
| 15 |
|
| 16 |
+
## Open-source quantization
|
| 17 |
+
|
| 18 |
+
### Qwen4-27B NVFP4 — preparation stage
|
| 19 |
+
|
| 20 |
+
We are preparing a reproducible NVFP4 quantization and release workflow for the anticipated Qwen4-27B model. The toolkit, CI, artifact checks, and release checklist are available on GitHub.
|
| 21 |
+
|
| 22 |
+
**Current status: preparation only.** No Qwen4 model weights, benchmark results, or validated Qwen4 architecture compatibility are available yet. Any future HF checkpoint release requires confirmation of the official source and license, GPU inference tests, and quality evaluation. This is an independent community project, not affiliated with Qwen or NVIDIA.
|
| 23 |
+
|
| 24 |
+
- [GitHub project](https://github.com/IteraCompute/qwen4-27b-nvfp4)
|
| 25 |
+
- [Quantization and release checklist](https://github.com/IteraCompute/qwen4-27b-nvfp4/blob/main/docs/RELEASE.md)
|
| 26 |
+
- [Release progress](https://github.com/IteraCompute/qwen4-27b-nvfp4/issues/2)
|
| 27 |
+
|
| 28 |
+
中文:目前已开源量化工具和发布流程;模型权重将在官方模型发布并完成 GPU 与质量验证后准备上线。
|
| 29 |
+
|
| 30 |
## Hugging Face Inference Provider
|
| 31 |
|
| 32 |
Our Hugging Face Inference Provider integration is currently being prepared for review. Initial conversational model mappings include:
|
|
|
|
| 41 |
## Resources
|
| 42 |
|
| 43 |
- [Website](https://iteracompute.com/)
|
| 44 |
+
- [GitHub organization](https://github.com/IteraCompute)
|
| 45 |
- [API documentation](https://iteracompute.com/docs.html)
|
| 46 |
- [Privacy policy](https://iteracompute.com/privacy.html)
|
| 47 |
- Provider partnerships: [partnerships@iteracompute.com](mailto:partnerships@iteracompute.com)
|