Skip to content

convert: text-only support for GLM-4.1V-9B-Thinking #14823

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Jul 23, 2025
Merged

convert: text-only support for GLM-4.1V-9B-Thinking #14823

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 10 additions & 2 deletions convert_hf_to_gguf.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6486,7 +6486,7 @@ def prepare_tensors(self):
self.gguf_writer.add_max_alibi_bias(self.max_alibi_bias)


@ModelBase.register("Glm4ForCausalLM")
@ModelBase.register("Glm4ForCausalLM", "Glm4vForConditionalGeneration")
class Glm4Model(TextModel):
model_arch = gguf.MODEL_ARCH.GLM4

Expand All @@ -6508,14 +6508,22 @@ def set_vocab(self):

def set_gguf_parameters(self):
super().set_gguf_parameters()
rope_dim = self.hparams["head_dim"]
if (rope_dim := self.hparams.get("head_dim")) is None:
rope_dim = self.hparams["hidden_size"] // self.hparams["num_attention_heads"]
self.gguf_writer.add_rope_dimension_count(int(rope_dim * self.hparams.get("partial_rotary_factor", 0.5)))
rope_scaling = self.hparams.get("rope_scaling") or {}
if rope_scaling.get("rope_type", rope_scaling.get("type")) == "yarn" and "factor" in rope_scaling:
self.gguf_writer.add_rope_scaling_type(gguf.RopeScalingType.YARN)
self.gguf_writer.add_rope_scaling_factor(rope_scaling["factor"])
self.gguf_writer.add_rope_scaling_orig_ctx_len(rope_scaling["original_max_position_embeddings"])

def modify_tensors(self, data_torch: Tensor, name: str, bid: int | None) -> Iterable[tuple[str, Tensor]]:
if name.startswith("model.visual."): # ignore visual part of Glm4v
return []
elif name.startswith("model.language_model."):
name = name.replace("language_model.", "") # for Glm4v
return super().modify_tensors(data_torch, name, bid)


@ModelBase.register("GlmForCausalLM", "ChatGLMModel", "ChatGLMForConditionalGeneration")
class ChatGLMModel(TextModel):
Expand Down