AI가 움직이고 미래가 업데이트됩니다.Qwen3.8-Flash-Next-MTP-GGUF
The multi-token-prediction head that ships inside Qwen/Qwen3.8-Flash-Next: one full `qwen4exp` block (attention + 512-expert MoE + hyper-connections, 4B parameters), trained jointly with the model. The upstream converter (PR 27742) drops it, so the popular GGUFs have none. This one is exported straight from the bf16 checkpoint and quantized with plain `llama-quantize`, so it runs on stock llama.cpp kernels (any backend), not a fork's quant types. **This is a draft head, not a model.** Use it as `-md` next to any Flash-Next GGUF target (tested with unsloth's UD-Q4_K_XL and UD-IQ4_XS). Set `LLAMA_ATTN_ROT_DISABLE=1`: upstream's quantized-KV activation rotation (#21038) is not supported by the qwen4exp attention path and the server aborts at load without it.
번역 준비 중입니다. 원문을 표시합니다.
The multi-token-prediction head that ships inside Qwen/Qwen3.8-Flash-Next: one full `qwen4exp` block (attention + 512-expert MoE + hyper-connections, 4B parameters), trained jointly with the model. The upstream converter (PR 27742) drops it, so the popular GGUFs have none. This one is exported straight from the bf16 checkpoint and quantized with plain `llama-quantize`, so it runs on stock llama.cpp kernels (any backend), not a fork's quant types. **This is a draft head, not a model.** Use it as `-md` next to any Flash-Next GGUF target (tested with unsloth's UD-Q4_K_XL and UD-IQ4_XS). Set `LLAMA_ATTN_ROT_DISABLE=1`: upstream's quantized-KV activation rotation (#21038) is not supported by the qwen4exp attention path and the server aborts at load without it.