sleepy/llama.cpp
1
0
Fork 0
Code Issues 11 Pull Requests Actions 109 Packages Projects Releases Wiki Activity

mtmd, llama : Update HunyuanVL vision-language model support (#22037)

Browse Source 
* mtmd, llama : add HunyuanVL vision-language model support

- add LLM_ARCH_HUNYUAN_VL with M-RoPE (XD-RoPE) support
- add PROJECTOR_TYPE_HUNYUANVL with PatchMerger vision encoder
- add HunyuanVL-specific M-RoPE position encoding for image tokens
- add GGUF conversion for HunyuanVL vision and text models
- add smoke test in tools/mtmd/tests.sh

* fix: fix HunyuanVL XD-RoPE h/w section order

* fix: Remove redundant code

* convert : fix HunyuanOCR / HunyuanVL conversion
 - Tested locally: both HunyuanOCR and HunyuanVL-4B convert to GGUF
 - successfully and produce correct inference output on Metal (F16 / Q8_0).

* clip : fix -Werror=misleading-indentation in bilinear resize

* fix CI: convert_hf_to_gguf type check error
 - convert_hf_to_gguf.py: give HunyuanVLTextModel.__init__ an explicit `dir_model: Path` parameter so ty can infer the type for load_hparams instead of reporting `Unknown | None`.

---------

Co-authored-by: wendadawen <wendadawen@tencent.com>
This commit is contained in:
manayang
2026-04-22 17:58:43 +08:00
committed by GitHub
parent 750579ff14
commit 7bfe60fdf9
13 changed files with 336 additions and 27 deletions
Download Patch File Download Diff File Expand all files Collapse all files
gguf-py/gguf/constants.py
+20
View File
@@ -197,6 +197,7 @@ class Keys:
FREQ_BASE_SWA = "{arch}.rope.freq_base_swa"
SCALING_TYPE = "{arch}.rope.scaling.type"
SCALING_FACTOR = "{arch}.rope.scaling.factor"
SCALING_ALPHA = "{arch}.rope.scaling.alpha"
SCALING_ATTN_FACTOR = "{arch}.rope.scaling.attn_factor"
SCALING_ORIG_CTX_LEN = "{arch}.rope.scaling.original_context_length"
SCALING_FINETUNED = "{arch}.rope.scaling.finetuned"
@@ -471,6 +472,7 @@ class MODEL_ARCH(IntEnum):
ERNIE4_5_MOE = auto()
HUNYUAN_MOE = auto()
HUNYUAN_DENSE = auto()
HUNYUAN_VL = auto()
SMOLLM3 = auto()
GPT_OSS = auto()
LFM2 = auto()
@@ -957,6 +959,7 @@ MODEL_ARCH_NAMES: dict[MODEL_ARCH, str] = {
MODEL_ARCH.FALCON_H1: "falcon-h1",
MODEL_ARCH.HUNYUAN_MOE: "hunyuan-moe",
MODEL_ARCH.HUNYUAN_DENSE: "hunyuan-dense",
MODEL_ARCH.HUNYUAN_VL: "hunyuan_vl",
MODEL_ARCH.SMOLLM3: "smollm3",
MODEL_ARCH.GPT_OSS: "gpt-oss",
MODEL_ARCH.LFM2: "lfm2",
@@ -3489,6 +3492,22 @@ MODEL_TENSORS: dict[MODEL_ARCH, list[MODEL_TENSOR]] = {
MODEL_TENSOR.FFN_DOWN,
MODEL_TENSOR.FFN_UP,
],
MODEL_ARCH.HUNYUAN_VL: [
MODEL_TENSOR.TOKEN_EMBD,
MODEL_TENSOR.OUTPUT_NORM,
MODEL_TENSOR.OUTPUT,
MODEL_TENSOR.ATTN_NORM,
MODEL_TENSOR.ATTN_Q,
MODEL_TENSOR.ATTN_Q_NORM,
MODEL_TENSOR.ATTN_K,
MODEL_TENSOR.ATTN_K_NORM,
MODEL_TENSOR.ATTN_V,
MODEL_TENSOR.ATTN_OUT,
MODEL_TENSOR.FFN_NORM,
MODEL_TENSOR.FFN_GATE,
MODEL_TENSOR.FFN_DOWN,
MODEL_TENSOR.FFN_UP,
],
MODEL_ARCH.SMOLLM3: [
MODEL_TENSOR.TOKEN_EMBD,
MODEL_TENSOR.OUTPUT_NORM,
@@ -4138,6 +4157,7 @@ class VisionProjectorType:
YOUTUVL = "youtuvl"
NEMOTRON_V2_VL = "nemotron_v2_vl"
HUNYUANOCR = "hunyuanocr"
HUNYUANVL = "hunyuanvl"
# Items here are (block size, type size)
gguf-py/gguf/gguf_writer.py
+3
View File
@@ -973,6 +973,9 @@ class GGUFWriter:
def add_rope_scaling_factor(self, value: float) -> None:
self.add_float32(Keys.Rope.SCALING_FACTOR.format(arch=self.arch), value)
def add_rope_scaling_alpha(self, value: float) -> None:
self.add_float32(Keys.Rope.SCALING_ALPHA.format(arch=self.arch), value)
def add_rope_scaling_attn_factors(self, value: float) -> None:
self.add_float32(Keys.Rope.SCALING_ATTN_FACTOR.format(arch=self.arch), value)