Downloads · 30 days
48
5% of all-time downloads
jakeatx/slimder-qwen38-ream288-depth32-agentic-ngram50-mixed-nvfp4-v2
slimder-qwen38-ream288-depth32-agentic-ngram50-mixed-nvfp4-v2 is a text generation model from jakeatx. Use it when you need the model to write or continue text. The card lists the license as apache-2.0.
This is the corrected mixed-precision runtime checkpoint derived from the promoted 50%-compact REAM-288 checkpoint. Expert tensors are stored in NVFP4, while the compact PLE n-gram table is stored in FP8 with per-row…
Downloads · 30 days
48
5% of all-time downloads
All-time downloads
937
Public
Parameters
56.5B
64.2 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors64.2 GB · 100%
How the weights are stored.
F8_E4M328.3B · 50%
From the Hugging Face model README
This is the corrected mixed-precision runtime checkpoint derived from the promoted 50%-compact REAM-288 checkpoint. Expert tensors are stored in NVFP4, while the compact PLE n-gram table is stored in FP8 with per-row FP16 scales.
The v2 exporter uses one shared global scale for each fused gate/up expert tensor, matching the vLLM Marlin loader's contract. This corrects the scale mismatch present in the superseded first mixed export.
24a0cf8360699362b6a99da498a4138b6a390918bf9cfc9df5b15bcdc3397983This model uses the experimental qwen4_exp architecture, a compact PLE remap,
and FP8 PLE storage. Use the included custom model/runtime files and a vLLM
build with the required Qwen4 experimental and Marlin NVFP4 support. The
checkpoint is intended for research and recovery fine-tuning, not as a generic
drop-in Transformers model.
The recovery QLoRA adapter and its held-out validation evidence are published separately after the guarded recovery run completes.
sjakek/slimder-qwen38-ream288-depth32-agenticsjakek/slimder-qwen38-ream288-depth32-agentic-ngram50-compactsjakek/slimder-qwen38-ngram-activation-refine-20260901The runtime and sealed corpus gates passed, but this remains an experimental pruned and quantized checkpoint. Broad task-level evaluation is still required before production use.
<!-- qwen38-perian-lineage:start -->This repository is retained in the Qwen3.8 Perian checkpoints collection. Its exact position in the lineage is: Mixed-precision training base after depth, expert-width, and PLE n-gram reductions, before the final rank-32 QLoRA.
The final Qwen3.8 Perian GGUF release combines three reductions and one post-training stage:
Earlier checkpoints in this collection do not inherit later stages merely by being listed beside them; the stage statement above is authoritative for this artifact.
<!-- qwen38-perian-lineage:end -->