Text Generation
memra
GGUF
nvfp4
speculative-decoding
mtp
conversational
blackwell
qwen3
Eval Results (legacy)
Instructions to use tiyuvta/Qwen3.8-27B-NVFP4-MTP-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- memra
How to use tiyuvta/Qwen3.8-27B-NVFP4-MTP-GGUF with memra:
# memra serves NVIDIA Blackwell workstation and consumer cards (sm_120a), with a # compile-gated Hopper lane. Prebuilt binaries need Linux x86_64 and driver 580+, # and no CUDA toolkit. curl -fsSL https://raw.githubusercontent.com/avifenesh/memra/main/tools/install.sh | sh
# One chat-templated generation. In a repo with several GGUF files, append # :<substring> to choose one, for example hf:tiyuvta/Qwen3.8-27B-NVFP4-MTP-GGUF:Q4_K_M MEMRA_CHAT=1 run-gen hf:tiyuvta/Qwen3.8-27B-NVFP4-MTP-GGUF --prompt "Explain KV caches in one sentence."
# Or an OpenAI-compatible server on 127.0.0.1:8080. MEMRA_MODELS="model=hf:tiyuvta/Qwen3.8-27B-NVFP4-MTP-GGUF" memra-server
- Notebooks
- Google Colab
- Kaggle
Upload mtp-Qwen3.8-27B-NVFP4-frspec-mixed32768.gguf with huggingface_hub
Browse files
.gitattributes
CHANGED
|
@@ -38,3 +38,4 @@ mtp-Qwen3.8-27B-NVFP4-frspec-sxc32768.gguf filter=lfs diff=lfs merge=lfs -text
|
|
| 38 |
q38-ranks-sxc32768.gguf filter=lfs diff=lfs merge=lfs -text
|
| 39 |
q38-ranks-prose-32768.gguf filter=lfs diff=lfs merge=lfs -text
|
| 40 |
mtp-Qwen3.8-27B-NVFP4-frspec-prose32768.gguf filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 38 |
q38-ranks-sxc32768.gguf filter=lfs diff=lfs merge=lfs -text
|
| 39 |
q38-ranks-prose-32768.gguf filter=lfs diff=lfs merge=lfs -text
|
| 40 |
mtp-Qwen3.8-27B-NVFP4-frspec-prose32768.gguf filter=lfs diff=lfs merge=lfs -text
|
| 41 |
+
mtp-Qwen3.8-27B-NVFP4-frspec-mixed32768.gguf filter=lfs diff=lfs merge=lfs -text
|
mtp-Qwen3.8-27B-NVFP4-frspec-mixed32768.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a47848be6f918b93f9647b67b302382c750774e83907d3cc7b6f2c42057b8e07
|
| 3 |
+
size 1242869536
|