Skip to content

add the configs for qwen3-vl-8b-instruct model #542

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

Open
sunny-infra wants to merge 3 commits into
base: main
Choose a base branch
from
Open

add the configs for qwen3-vl-8b-instruct model #542

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
41ba68b
add the configs for qwen3-vl-8b-instruct model
sunny-infra Apr 23, 2026
03c15e7
Merge branch 'sgl-project:main' into main
sunny-infra Jun 11, 2026
6cf3eed
Merge branch 'sgl-project:main' into main
sunny-infra Jul 9, 2026
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
41 changes: 41 additions & 0 deletions configs/qwen3-vl-8b-instruct-eagle3.json
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,41 @@
{
"architectures": [
"LlamaForCausalLMEagle3"
],
"image_token_id": 151655,
"model_type": "llama",
"target_model_type": "qwen3_vl",

@gemini-code-assist gemini-code-assist Bot Apr 23, 2026

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

high

The target_model_type set to "qwen3_vl" is not currently supported in the AutoDistributedTargetModel._model_mapping within specforge/modeling/auto.py. Without adding this mapping (e.g., mapping to a Qwen3VLForCausalLM class), the target model will fail to instantiate during training or inference.

"attention_bias": false,
"attention_dropout": 0.0,
"bos_token_id": 151643,
"dtype": "bfloat16",
"eos_token_id": 151645,
"head_dim": 128,
"hidden_act": "silu",
"hidden_size": 4096,
"initializer_range": 0.02,
"intermediate_size": 12288,
"max_position_embeddings": 262144,
"num_attention_heads": 32,
"num_hidden_layers": 1,
"num_key_value_heads": 8,
"rms_norm_eps": 1e-06,
"rope_scaling": {
"mrope_interleaved": true,

@gemini-code-assist gemini-code-assist Bot Apr 23, 2026

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

The mrope_interleaved parameter is not currently handled by the apply_multimodal_rotary_pos_emb function in specforge/modeling/draft/llama3_eagle.py. The existing implementation assumes contiguous chunks for the temporal, height, and width sections based on mrope_section. If Qwen3-VL uses an interleaved layout, the draft model's rotary embeddings will be incorrect.

"mrope_section": [
24,
20,
20
],
"rope_type": "mrope"
},
"rope_theta": 5000000,

@gemini-code-assist gemini-code-assist Bot Apr 23, 2026

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

high

The rope_theta value of 5,000,000 will be ignored for the draft model. In specforge/modeling/draft/llama3_eagle.py, the LlamaAttention._init_rope method instantiates LlamaMutiRotaryEmbedding (used for mrope) without passing the base (theta) parameter, causing it to default to 10,000. This will lead to a significant mismatch between the draft and target model embeddings.

"use_cache": true,
"vocab_size": 151936,
"tie_word_embeddings": false,
"transformers_version": "4.57.0.dev0",
"video_token_id": 151656,
"vision_end_token_id": 151653,
"vision_start_token_id": 151652,
"draft_vocab_size": 32000
}