AngelSlim/Hy4-flash-GGUF
AI model by AngelSlim
From the publisher
Model card & documentation
Source previewRead the publisher’s intended use, setup instructions, evaluations and limitations. The original model card is the source of truth.
中文 | English ✒️ TechnicalReport | 📖 Documentation | 🤗 Hugging Face | 🤖 ModelScope 💬 WeChat | 🫨 Discord Five GGUF builds of Hy4-flash: tencent/Hy4-flash Hy4-flash-Q4KM.gguf — standard 4-bit, the safe default. The routed experts follow llama.cpp's Q4KM recipe: gate/up are Q4K, and down gets Q6K on 19 layers via llama.cpp's fixed rule and Q4K on the other 22. Most non-expert weights are Q80. Use this unless you are memory-constrained. Hy4-flash-SPARK-4Bit.gguf — the SPARK-4Bit build (NVIDIA's 4-bit floating-point format). Suited for Spark devices. Hy4-flash-SPARK-4Bit-Fast.gguf — a variant of the SPARK-4Bit build, 2.7 GiB smaller. Also suited for Spark devices. Hy4-flash-Q3K.gguf — mixed Q3K/Q4K experts. The routed-expert gate/up projections keep Q4K on 19 layers chosen by quantization sensitivity and drop to Q3K on the other 22; down is Q4K. Everything outside the routed experts is the same as Q4KM. Hy4-flash-IQ2-AH.gguf — the smallest of the five. The routed-expert gate/up…
Read the full model card ↗ · Preview checked 2026-11-19T09:40:24.752Z
Inside the original model card — Document outline
- A more accessible, comprehensive, and efficient toolkit for large model compression.
- Hy4-flash GGUF
- 1. What these are
- 2. SPEED
- 3. How to use
Headings are captured from the source. Links open the publisher’s document, not a locally hosted copy.
Links from the model card
References supplied by the publisher, not independently verified endorsements. Check the destination before downloading files or entering credentials.
Documentation belongs to its respective authors. A listing is not a grant of reuse or training rights. Confirm the document’s own terms at the source.
Model overview
AngelSlim/Hy4-flash-GGUF is a task-unspecified model repository published by AngelSlim on Hugging Face. A library was not reported.
This page summarizes Hub metadata. For intended use, training data, evaluation results and limitations, consult the original model card.
Model facts
- Task
- Not reported
- Library
- Not reported
- Recent downloads (30 days)
- 744
- Cumulative likes
- 61
- Hugging Face trending score
- 54
- Reported safetensors parameters
- Not reported
- Architecture
- Not reported
- License
- Not reported — check the model card
- Access
- Not gated by Hugging Face
- Created
- 2026-11-16T14:28:43.000Z
- Last modified
- 2026-11-18T15:32:00.000Z
Compatibility and lineage
Reported languages: zh, en
Reported base models: tencent/Hy4-flash
Parameter count is not a RAM/VRAM requirement. Check precision, quantization, context length and runtime compatibility in the model card; no hardware or API-cost claim is inferred here.
Use and evaluate
Check license terms and access requirements first. Review the model card’s documented loading instructions, evaluations and safety limitations. Benchmark scores are not imported by this directory, and popularity is not an accuracy ranking.
No model weights are downloaded or executed by AltAPIs. Never enable remote model code without reviewing it.
Source and freshness
Source: Hugging Face Hub. Metadata observed 2026-11-19T07:00:11.321Z. Daily imports are snapshots, not real-time monitoring.
Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.