Card: the project is now TurboQwen (github.com/Ar4ikov/TurboQwen)

README.md

@@ -24,7 +24,7 @@ tags:
 
 [Qwen3.8-27B-Uncensored-AWQ-W4A16-ASYM](https://huggingface.co/Ar4ikov/Qwen3.8-27B-Uncensored-AWQ-W4A16-ASYM) prepared for
 [HyperQwen](https://github.com/syv-ai/HyperQwen) — Qwen3.8-27B served fast on one 24 GB card — the way the
-[vllm-hyprfastQwen](https://github.com/Ar4ikov/vllm-hyprfastQwen) image expects it. This is the **uncensored** (abliterated) finetune; its behaviour is inherited from [orcarouter/Qwen3.8-27B-Uncensored](https://huggingface.co/orcarouter/Qwen3.8-27B-Uncensored). The body is untouched: int4 asymmetric
+[TurboQwen](https://github.com/Ar4ikov/TurboQwen) image expects it. This is the **uncensored** (abliterated) finetune; its behaviour is inherited from [orcarouter/Qwen3.8-27B-Uncensored](https://huggingface.co/orcarouter/Qwen3.8-27B-Uncensored). The body is untouched: int4 asymmetric
 AWQ, group 128, zero points, quantized with llm-compressor from bf16 weights; the vision
 tower, the SSM gate projections and the MTP head's norms stay bf16. What changed is what
 HyperQwen's `prepare/` pipeline changes, so that a 24 GB card has room for a KV cache and
@@ -45,7 +45,7 @@ speculative decoding has something small to score:
 The container does everything (download, verify, serve), with the vision tower on:
 
 ```bash
-git clone https://github.com/Ar4ikov/vllm-hyprfastQwen && cd vllm-hyprfastQwen
+git clone https://github.com/Ar4ikov/TurboQwen && cd TurboQwen
 cp .env.example .env            # CHECKPOINT=uncensored
 docker compose --profile single up -d
 ```
@@ -76,7 +76,7 @@ real prompts with 1,024-token answers; tok/step = tokens accepted per forward pa
 
 Images are described correctly in every profile (a drawn red square, blue circle and a
 line of text: `boost/image_smoke.py`). The whole table, the int8 (W4A8) profiles and the
-kernel measurements: [github.com/Ar4ikov/vllm-hyprfastQwen](https://github.com/Ar4ikov/vllm-hyprfastQwen).
+kernel measurements: [github.com/Ar4ikov/TurboQwen](https://github.com/Ar4ikov/TurboQwen).
 
 ## Why a separate repo