AIが動く。未来が更新される。Qwen3.8-27B-NVFP4-RTX5090-LMHead4
> ## ⚠️ Superseded > > The main checkpoint now has the NVFP4 `lm_head` **and** the MTP head removed — it is strictly better than this variant (17.92 GB, 2 shards). > > **Use `Qwen3.8-27B-NVFP4-RTX5090` instead.** This repo stays up so existing pins keep working. `Qwen3.8-27B-NVFP4-RTX5090` with **`lm_head` quantized to NVFP4** instead of left in BF16. Every other tensor is bit-identical to the parent. At concurrency 1 this model is **weight-bandwidth bound, not compute bound**. Measured on the parent: 18.80 GiB of resident weights at 81.6 tok/s implies a **1.65 TB/s** read rate against the RTX 5090's 1.79 TB/s spec — about **92% of peak**. Decode is streaming the whole model once per token.
翻訳準備中のため原文を表示しています
> ## ⚠️ Superseded > > The main checkpoint now has the NVFP4 `lm_head` **and** the MTP head removed — it is strictly better than this variant (17.92 GB, 2 shards). > > **Use `Qwen3.8-27B-NVFP4-RTX5090` instead.** This repo stays up so existing pins keep working. `Qwen3.8-27B-NVFP4-RTX5090` with **`lm_head` quantized to NVFP4** instead of left in BF16. Every other tensor is bit-identical to the parent. At concurrency 1 this model is **weight-bandwidth bound, not compute bound**. Measured on the parent: 18.80 GiB of resident weights at 81.6 tok/s implies a **1.65 TB/s** read rate against the RTX 5090's 1.79 TB/s spec — about **92% of peak**. Decode is streaming the whole model once per token.