Hy4-preview GGUF

Three GGUF builds of Hy4-Preview: https://huggingface.co/tencent/Hy4-preview

Language / 语言: English · 中文

| file | size | bpw | notes |

|---|---:|---:|---|

| Hy4-preview-Q4_K_M.gguf | 435.20 GiB | 4.86 | standard 4-bit, safe default |

|Hy4-preview-UD-IQ1_M.gguf| 219.83 GiB| 2.44 | Using the UD-IQ1_M quantization strategy, half the size |

| Hy4-preview-STQ1_0.gguf | 213.66 GiB | 2.38 | Using the MIX_STQ1_0 quantization strategy, half the size |

Neither file runs on stock llama.cpp. The hyv4 architecture is not upstream. Apply the

patches in hy4-preview-patch/

---

<a name="english"></a>

English

1. What these are

Hy4-preview-Q4_K_M.gguf — a conventional Q4_K_M. Most tensors are Q4_K; ffn_down_exps

gets Q6_K on 37 layers via llama.cpp's own logic. Use this unless you are memory-constrained.

Hy4-preview-UD-IQ1_M.gguf - mixed precision with UD-IQ1_M strategy at ~2.44 bpw, roughly half the size for the

same model. The routed-expert gate/up projections run at 1.75 bpw (IQ1_M) and 2.0625 bpw (IQ2_XXS).

Hy4-preview-STQ1_0.gguf — mixed precision with MIX-STQ1_0 strategy at ~2.38 bpw, roughly half the size for the

same model. The routed-expert gate/up projections run at 1.3125 bpw (STQ1_0) on 29 layers and

2.0625 bpw (IQ2_XXS) on the other 48. See section 3.

2. Running them

Build a patched llama.cpp

git clone https://github.com/ggml-org/llama.cpp && cd llama.cpp
git checkout 0cea36222

git apply hy4-preview-patch/0001-hyv4-architecture.patch
git apply hy4-preview-patch/0002-stq1_0-quant-and-cuda.patch   # skip if only using Q4_K_M

export PATH=/usr/local/cuda-13.0/bin:$PATH CUDACXX=/usr/local/cuda-13.0/bin/nvcc
cmake -B build-cuda -DGGML_CUDA=ON -DLLAMA_CURL=OFF -DGGML_NATIVE=OFF \
      -DCMAKE_BUILD_TYPE=Release -DCMAKE_CUDA_ARCHITECTURES=90 \
      -DLLAMA_BUILD_UI=OFF -DLLAMA_USE_PREBUILT_UI=OFF
cmake --build build-cuda --target llama-cli llama-bench llama-quantize -j 48

Set -DCMAKE_CUDA_ARCHITECTURES for your GPU (90 = H20/H100). Both -DLLAMA_BUILD_UI=OFF and

-DLLAMA_USE_PREBUILT_UI=OFF are needed for an offline build; the first alone still downloads

prebuilt assets.

Then

# single prompt
build-cuda/bin/llama-cli -m Hy4-preview-Q4_K_M.gguf -ngl 99 -c 8192 \
    --temp 0 -n 512 --no-warmup --jinja -st -f prompt.txt

# throughput
build-cuda/bin/llama-bench -m Hy4-preview-STQ1_0.gguf -ngl 99 -p 512 -n 128 -r 3

• --jinja is required for chat. The HY4 chat template matches no llama.cpp built-in family.
• Keep the GGUF on local disk. llama.cpp mmaps weights; over NFS random page faults run at

~12 MB/s, turning a 1-minute load into hours.

• Use -st -f prompt.txt for a single prompt. -no-cnv is ignored in this build and it will

spin printing > on EOF.

• VRAM for full residency: ~435 GiB (Q4_K_M) or ~214 GiB (STQ1_0). With less, lower -ngl.

Measured on 8x H20:

| | prefill (pp512) | decode (tg128) |

|---|---:|---:|

| STQ1_0 | 204.56 ± 1.42 t/s | 20.47 ± 0.02 t/s |

Python tools reading these files must use the patched gguf-py with an absolute path:

sys.path.insert(0, '/path/to/llama.cpp/gguf-py').

3. STQ1_0 and the mixed-precision strategy

The format. STQ1_0 comes from llama.cpp PR #22836. Weights are ternary {-d, 0, +d}, with

exactly one of every four lanes forced to zero (3:4 sparsity). Each group of 4 weights is a

4-bit code plus a 1-bit table-select, indexing a 32-entry codebook; one fp16 scale covers 256

weights. That is 2 + 32 + 8 = 42 bytes per 256 weights = 1.3125 bpw.

Our encoder. Upstream's quantizer targets QAT inputs already on the ternary grid: it ignores

the imatrix, sets d = amax, and zeroes argmin |x|. That is weak for post-training

quantization. We keep the format byte-identical and change only two decisions:

1. Weighted least-squares scale, d = sum(wselx) / sum(w*sel^2) instead of d = amax.
2. Imatrix-aware zero placement — zero the lane minimising w[j]*(x[j]^2 - (|x[j]|-d)^2),

the incremental cost rather than the smallest magnitude.

alternating for 3 rounds. Measured on 1200 real expert rows: the LS scale alone gives -89.7%

weighted SSD, and the imatrix terms a further -4.1% of the remainder. The headline win is the

scale — amax pins d to the single largest outlier among 256 weights.

Where the bits go. The three routed-expert families are 97.7% of all parameters, so the

recipe spends freely on everything else:

| family | STQ1_0 build | why |

|---|---|---|

| ffn_gate_exps / ffn_up_exps | STQ1_0 (29 layers) / IQ2_XXS (48 layers) | the bulk; layer choice is imatrix-derived |

| ffn_down_exps | IQ3_XXS, IQ4_XS on last 3 | writes straight into the residual stream, so its error is not attenuated by a later gate — deliberately 2 levels higher |

| attention out / gate / q_a | Q5_K | llama.cpp only auto-bumps these when n_expert == 8; HY4 has 256 |

| MLA q_b/k_b/v_b/kv_a_mqa | Q8_0 | HY4's split names miss llama.cpp's substring match, so they get no automatic bump |

| DSA indexer | Q8_0 / F32 | 105 tensors, 0.21 GiB total, gates which 2048 tokens each query sees |

| iHC *_fn, router, norms, sink | F32 | mirrors the reference's _keep_in_fp32_modules |

| output (lm_head) | F32 | via --leave-output-tensor |

4. Building a runtime

Re-quantizing from bf16

The recipe files are included. An imatrix is mandatory for STQ1_0 — its encoder uses it for

the scale solve and zero placement.

build-cuda/bin/llama-quantize --dry-run --imatrix imatrix.gguf \
    --tensor-type-file Hy4-preview-STQ1_0.tensortypes --leave-output-tensor \
    HY4.bf16.gguf out.gguf IQ1_M          # Q4_K_M build: use Q4_K_M as the base ftype

---

<a name="中文"></a>

中文

| file | size | bpw | notes |

|---|---:|---:|---|

| Hy4-preview-Q4_K_M.gguf | 435.20 GiB | 4.86 | 标准4bit gguf |

|Hy4-preview-UD-IQ1_M.gguf| 219.83 GiB| 2.44 | 使用 UD-IQ1_M 混合精度量化压缩 |

| Hy4-preview-STQ1_0.gguf | 213.66 GiB | 2.38 | 使用 MIX-STQ1_0 混合精度量化压缩 |

1. 这是什么

Hy4-preview-Q4_K_M.gguf —— 常规 Q4_K_M。多数张量为 Q4_K，ffn_down_exps 由 llama.cpp

自身逻辑提到 Q6_K（37 层）。没有显存压力就用这个。

Hy4-preview-STQ1_0.gguf —— 约 2.44 bpw 的混合精度，使用 UD-IQ1_M 混合精度量化压缩, 同一个模型体积减半。

Hy4-preview-STQ1_0.gguf —— 约 2.38 bpw 的混合精度，使用 MIX-STQ1_0 混合精度量化压缩, 同一个模型体积减半。路由专家的

gate/up 在 29 层用 1.3125 bpw（STQ1_0），另 48 层用 2.0625 bpw（IQ2_XXS）。见第 3 节。

2. 如何使用

git clone https://github.com/ggml-org/llama.cpp && cd llama.cpp
git checkout 0cea36222

git apply hy4-preview-patch/0001-hyv4-architecture.patch
git apply hy4-preview-patch/0002-stq1_0-quant-and-cuda.patch   # 只用 Q4_K_M 可跳过

export PATH=/usr/local/cuda-13.0/bin:$PATH CUDACXX=/usr/local/cuda-13.0/bin/nvcc
cmake -B build-cuda -DGGML_CUDA=ON -DLLAMA_CURL=OFF -DGGML_NATIVE=OFF \
      -DCMAKE_BUILD_TYPE=Release -DCMAKE_CUDA_ARCHITECTURES=90 \
      -DLLAMA_BUILD_UI=OFF -DLLAMA_USE_PREBUILT_UI=OFF
cmake --build build-cuda --target llama-cli llama-bench llama-quantize -j 48

-DCMAKE_CUDA_ARCHITECTURES 按自己的 GPU 设置（90 = H20/H100）。离线构建同时需要

-DLLAMA_BUILD_UI=OFF 与 -DLLAMA_USE_PREBUILT_UI=OFF，只给前者仍会去下载预构建资源。

# 单条 prompt
build-cuda/bin/llama-cli -m Hy4-preview-Q4_K_M.gguf -ngl 99 -c 8192 \
    --temp 0 -n 512 --no-warmup --jinja -st -f prompt.txt

# 测速
build-cuda/bin/llama-bench -m Hy4-preview-STQ1_0.gguf -ngl 99 -p 512 -n 128 -r 3

• chat 必须加 --jinja。 HY4 的 chat template 不匹配 llama.cpp 任何内置模板家族。
• GGUF 必须放本地盘。 llama.cpp 用 mmap，NFS 随机页错误约 12 MB/s，本来 1 分钟的加载会变

成几小时。

• 单条 prompt 用 -st -f prompt.txt。 本 build 忽略 -no-cnv，遇 EOF 会一直打印 >。
• 全量驻留显存需求：约 435 GiB（Q4_K_M）或约 214 GiB（STQ1_0）。不够就降低 -ngl。

在 8 x H20 上实测（已确认 GPU 空闲、权重全驻显存）：

| | 预填充 (pp512) | 解码 (tg128) |

|---|---:|---:|

| STQ1_0 | 204.56 ± 1.42 t/s | 19.52 ± 0.01 t/s |

读这些文件的 Python 工具必须用打过补丁的 gguf-py，且用绝对路径：

sys.path.insert(0, '/path/to/llama.cpp/gguf-py')。

3. STQ1_0 与混合精度策略

格式。 STQ1_0 来自 llama.cpp PR #22836。权重为三值 {-d, 0, +d}，且**每 4 个 lane 强制

一个为零**（3:4 稀疏）。每 4 个权重存成 4-bit code 加 1-bit 选表位，索引一张 32 项码本；每

256 个权重共用一个 fp16 scale。即每 256 权重 2 + 32 + 8 = 42 字节 = 1.3125 bpw。

我们的编码器。 上游的量化器面向已落在三值网格上的 QAT 输入：直接忽略 imatrix，取

d = amax，并把零放在 argmin |x|。这对训练后量化（PTQ）很弱。我们保持格式逐字节一致，

只改两个决策：

1. 加权最小二乘 scale：d = sum(wselx) / sum(w*sel^2)，取代 d = amax。
2. imatrix-aware 零位置：零掉使 w[j]*(x[j]^2 - (|x[j]|-d)^2) 最小的 lane，即比较增量

代价，而非单纯的最小幅值。

两者交替 3 轮。在 1200 行真实专家权重上实测：仅最小二乘 scale 就带来 -89.7% 加权 SSD，

imatrix 项在残差上再补 -4.1%。主要收益来自 scale——amax 会把 d 钉在 256 个权重里

的单个最大离群值上。

bit 花在哪。 三个路由专家族占全部参数的 97.7%，所以配方在其余张量上舍得花：

| 张量族 | STQ1_0 产物 | 原因 |

|---|---|---|

| ffn_gate_exps / ffn_up_exps | STQ1_0（29 层）/ IQ2_XXS（48 层）| 体积主体；选层由 imatrix 推导 |

| ffn_down_exps | IQ3_XXS，最后 3 层 IQ4_XS | 直接写回残差流，误差不会被后续 gate 衰减，故刻意高两档 |

| attention out / gate / q_a | Q5_K | llama.cpp 只在 n_expert == 8 时自动提档，而 HY4 有 256 个专家 |

| MLA q_b/k_b/v_b/kv_a_mqa | Q8_0 | HY4 的拆分命名匹配不上 llama.cpp 的子串规则，完全拿不到自动提档 |

| DSA indexer | Q8_0 / F32 | 105 个张量共 0.21 GiB，却是决定每个 query 能看到哪 2048 个 token 的闸门 |

| iHC *_fn、router、norms、sink | F32 | 对齐参考实现的 _keep_in_fp32_modules |

| output（lm_head）| F32 | 通过 --leave-output-tensor |

从 bf16 重新量化

配方文件已随附。STQ1_0 强制需要 imatrix——它的编码器要用 imatrix 做 scale 求解与零位置选择。

build-cuda/bin/llama-quantize --dry-run --imatrix imatrix.gguf \
    --tensor-type-file Hy4-preview-STQ1_0.tensortypes --leave-output-tensor \
    HY4.bf16.gguf out.gguf IQ1_M          # Q4_K_M 产物：基础 ftype 用 Q4_K_M

---

Files

hy4-preview-patch/
  0001-hyv4-architecture.patch        18 files, +1632/-3   both GGUFs need this
  0002-stq1_0-quant-and-cuda.patch    25 files, +683/-4    STQ1_0 only
  Hy4-preview-STQ1_0.tensortypes      the STQ1_0 recipe
  Hy4-preview-Q4_K_M.tensortypes      the Q4_K_M recipe