Commit History

Fix model card: remove enforce-eager (fixed in vLLM 0.18.0 PR #35256), add benchmarks
6415a20
verified

raydelossantos commited on

Update model card: VLM architecture, enable-prefix-caching, weight structure docs
e256acb
verified

raydelossantos commited on

Remove old 3-shard files (replaced by single model.safetensors)
61950f0
verified

raydelossantos commited on

Fix weight keys for VLM: rename model.* to model.language_model.*, add visual encoder from base model
e2ec2d9
verified

raydelossantos commited on

Fix config.json: use VLM structure (model_type=qwen3_5, Qwen3_5ForConditionalGeneration) with quantization_config
7c83058
verified

raydelossantos commited on

Add vision_config to config.json for vLLM VLM backbone compat
a215e82
verified

raydelossantos commited on

Add processor_config.json with video processor for vLLM compat
e49380d
verified

raydelossantos commited on

Add preprocessor_config.json for vLLM compatibility (text-only quant)
1f34a7e
verified

raydelossantos commited on

Delete preprocessor_config.json with huggingface_hub
5e3e47d
verified

raydelossantos commited on

Upload config.json with huggingface_hub
03bfdf5
verified

raydelossantos commited on

Upload preprocessor_config.json with huggingface_hub
a304df7
verified

raydelossantos commited on

Upload config.json with huggingface_hub
0bd9a9d
verified

raydelossantos commited on

Upload folder using huggingface_hub
a9fc8f6
verified

raydelossantos commited on