Tag

Qwen3.8-27B-Distillation-40K

Contains 40,000 teacher-generated reasoning traces distilled from the Qwen3.8-27B model for supervised fine-tuning and analysis. Covers code, math, science and logic; each example pairs a <think> chain-of-thought with a final response and is distributed in JSONL/Parquet for SFT workflows.

Qwen3.8-Flash-Next-FP8

Provides FP8-quantized Hugging Face weights and config for Qwen3.8-Flash-Next (block size 128), preserving near-original performance. Compatible with Transformers, vLLM, SGLang and TokenSpeed; intended for efficient deployment of a 125B multimodal causal LM with very long context support.

GLM-5.3

A large open-weights MoE language model for complex coding, long-horizon agentic workflows, and cyber/security evaluations; post-trained from the GLM-5 family with substantial gains over GLM-5.2. Provides FP8/BF16 checkpoints and native support for very long contexts (up to 1M tokens).

GLM-5.3-Flash

A natively multimodal model for text and image→text generation, long-context reasoning, and complex coding/agent workloads. Uses 320B total / 18B active params with a hybrid sparse+linear attention and manifold-constrained hyper-connections to reduce long-context serving cost.

Hy4 preview

A 770B-parameter Mixture-of-Experts instruct model from Tencent that natively supports 1,048,576-token contexts, Gated DSA attention, and speculative MTP decoding; open-sourced under Apache-2.0 with BF16 and FP8 weights for deployable inference.

orcarouter/GLM-5.3-Flash-Uncensored-FP8

Drop-in abliterated (refusal-removed) build of GLM-5.3-Flash that bakes refusal-direction removal into block-FP8 safetensors, yielding an uncensored 320B (18B active) multimodal MoE model with a 1M-token context. Intended for red-teaming, interpretability, and robustness research; MIT license; not for production without added guardrails.

GLM 5.3 CRACK — Cybersecurity FP8

Provides a cybersecurity-focused CRACK variant of GLM-5.3 FP8 that reduces refusals for offensive-security, red-team, exploit-development and malware-analysis queries while retaining native FP8 speed on Hopper GPUs; MIT-licensed for authorized security work.

DeepSeek-V4-Flash-Vision-Exp

An experimental multimodal model that adds visual understanding to DeepSeek-V4-Flash: accepts text+image inputs and returns text analyses. Improves vision-dependent agent workflows while maintaining comparable text-only performance; released under an MIT license on Hugging Face.

nvidia/Qwen3.8-Flash-Next-NVFP4

NVFP4-quantized checkpoint of Qwen3.8-Flash-Next for GPU-optimized multimodal autoregressive inference — routed MoE experts in W4A4 NVFP4 while attention/ancillary layers remain BF16; ~2.7× smaller than the BF16 source and supports very long contexts.

VDN-Minimax-H3

Adds a plug-and-play linear-attention branch and LoRA adapters to MiniMax-H3 to run text-to-video generation faster than real-time (near-lossless quality tradeoffs). Includes an optimized FP8 inference stack and a community license with regional restrictions.

Why Gated DeltaNet Survives 4-Bit Quantization: NVFP4 W4A4 for the Recurrent Half of a Hybrid 27B LLM

Shows that fully quantizing all 496 linear layers—including the recurrent Gated DeltaNet—of a hybrid 27B LLM to 4-bit NVFP4 W4A4 preserves benchmark accuracy while reducing model size to 17.5 GiB and improving prefill speed; includes a calibrated NVFP4 checkpoint.

Swift-Qwen3.8-27B

A reasoning‑efficient fine-tune of Qwen3.8-27B that penalizes overthinking tokens to shorten internal reasoning traces — about 58.3% fewer thinking tokens with <1% accuracy loss and ~1.95× speedup; designed for long-context, multimodal and quantized deployments.