---
language:
- zh
- en
base_model:
- tencent/Hy4-flash
---
[中文](./README_CN.md) | English
### A more accessible, comprehensive, and efficient toolkit for large model compression.
✒️ [TechnicalReport](https://arxiv.org/abs/2602.21233) | 📖 [Documentation](https://angelslim.readthedocs.io/) | 🤗 [Hugging Face](https://huggingface.co/AngelSlim) | 🤖 [ModelScope](https://modelscope.cn/organization/AngelSlim)
💬 [WeChat](https://github.com/Tencent/AngelSlim/blob/main/docs/source/assets/angel_slim_wechat.png?raw=true) | 🫨 [Discord](https://discord.com/invite/dHVNeuNdFt)
# Hy4-flash GGUF
Five GGUF builds of Hy4-flash: [tencent/Hy4-flash](https://huggingface.co/tencent/Hy4-flash)
| file | model size | bpw (overall) | bpw (excl. Engram table) |
| -------------------------------- | ---------: | ------------: | -----------------------: |
| `Hy4-flash-Q4_K_M.gguf` | 156.79 GiB | 6.05 | 5.27 |
| `Hy4-flash-SPARK-4Bit.gguf` | 151.3 GiB | 5.84 | 4.78 |
| `Hy4-flash-SPARK-4Bit-Fast.gguf` | 148.6 GiB | 5.74 | 4.63 |
| `Hy4-flash-Q3_K.gguf` | 144.70 GiB | 5.58 | 4.62 |
| `Hy4-flash-IQ2-AH.gguf` | 125.73 GiB | 4.85 | 3.60 |

---
## 1. What these are
`Hy4-flash-Q4_K_M.gguf` — standard 4-bit, the safe default. The routed experts follow llama.cpp's
Q4_K_M recipe: `gate`/`up` are Q4_K, and `down` gets Q6_K on 19 layers via llama.cpp's fixed rule and
Q4_K on the other 22. Most non-expert weights are Q8_0. **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-Q3_K.gguf` — mixed Q3_K/Q4_K experts. The routed-expert `gate`/`up` projections keep
Q4_K on 19 layers chosen by quantization sensitivity and drop to Q3_K on the other 22; `down` is Q4_K.
Everything outside the routed experts is the same as Q4_K_M.
`Hy4-flash-IQ2-AH.gguf` — the smallest of the five. The routed-expert `gate`/`up` projections run
at 2.0625 bpw (IQ2_XXS) and `down` at Q4_K, while most non-expert weights stay in BF16 rather than Q8_0.
**Strata** version coming soon
## 2. SPEED
The test was conducted in real workbuddy scenarios and cases.
**Mac M5 Max 128GB**
| format | Prefill tok/s (4K / 16K / 32K) | Decode tok/s (4K / 16K / 32K) |
| ------ | ----------------------------: | ----------------------------: |
| Q4_K_M | 629.8 / 448.1 / 418.6 | 48.13 / 45.18 / 44.15 |
| Q3_K | 743.2 / 563.0 / 520.6 | 50.34 / 49.47 / 47.37 |
For code generation, Q3_K reaches up to 70 tok/s.
**RTX Spark 128GB**
| format | Prefill tok/s (4K / 16K / 32K) | Decode tok/s (4K / 16K / 32K) |
| --------------- | ----------------------------: | ----------------------------: |
| SPARK-4Bit-Fast | 799 / 717 / 707 | 31.95 / 33.14 / 33.39 |
| SPARK-4Bit | 765 / 686 / 666 | 25.80 / 28.16 / 30.3 |
For code generation, SPARK-4Bit-Fast reaches up to 40 tok/s.
## 3. How to use
**General Usage**
The support is in PR [#32517](https://github.com/ggml-org/llama.cpp/pull/32517) and not merged yet, so check out the PR branch:
```sh
git clone https://github.com/ggml-org/llama.cpp
cd llama.cpp
git fetch origin pull/32517/head:hy4-flash
git checkout hy4-flash
```
Build:
```sh
cmake -B build # add -DGGML_CUDA=ON for NVIDIA GPUs; Metal is on by default on macOS
cmake --build build --config Release -j
```
Chat in the terminal:
```sh
./build/bin/llama-cli -m Hy4-flash-Q4_K_M-00001-of-00002.gguf -c 8192 -ngl 99 --jinja
```
OpenAI-compatible server (http://localhost:8080):
```sh
./build/bin/llama-server -m Hy4-flash-Q4_K_M-00001-of-00002.gguf -c 8192 -ngl 99 --jinja --port 8080
```
Image input, with the mmproj file:
```sh
./build/bin/llama-mtmd-cli -m Hy4-flash-Q4_K_M-00001-of-00002.gguf --mmproj mmproj-Hy4-flash-bf16.gguf \
--image photo.png -p "Describe this image." -c 8192 -ngl 99 --jinja
```
**Mac M5 Max 128GB**
See [README_MAC-Metal.md](https://huggingface.co/AngelSlim/Hy4-flash-GGUF/blob/main/README_MAC-Metal.md)
**RTX Spark 128GB**
See [README_SPARK-4Bit.md](https://huggingface.co/AngelSlim/Hy4-flash-GGUF/blob/main/README_SPARK-4Bit.md)