Repository Issue Activity (beta)

#351 Add all available ONNX models to ORTConfigManager

#2379 qwen3-0.6 to onnx : index: 3 Got: 64 Expected: 128

#2474 export Qwen3-ASR-0.6B to onnx error

#488 Community contribution - `BetterTransformer` integration for more models!

#2488 Support text-classification task for gemma3_text models

#2487 `iostream error` when exporting models.

#2425 Add ONNX export support for Qwen3.5/3.6 MoE (qwen3_5_moe) architecture

#1777 DummyTextInputGenerator doesn't take into account type_vocab_size model config parameter

#2188 no __version__ attribute

#2431 ONNX export support for SmolVLM2 architecture (smolvlm2)

#2436 [Bug] Gemma-4 ONNX Export fails due to Mixed Head Dimensions (RoPE corruption) and MemoryError during validation for Qualcomm NPU via ONNX-Runtime

#2433 Support hunyuan_v1_dense architecture

#555 Community contribution - `optimum.exporters.onnx` support for new models!

#2438 Exporting Models: torch_dtype is deprecated

#2351 ONNX export fails for Qwen/Qwen3-4B-Thinking-2507 with RuntimeError

#2215 not able to convert DeepSeek-R1 into Onnx using optimum-cli

#2312 Exporting Mistral to ONNX Model

#2434 I follow the Documentation and I still can't fully export an ASR model to ONNX :(

#2338 RuntimeError converting Whisper model

#2412 the task text-classification support in the ONNX export for gemma3_text.

#2421 [Feature Request] Add ONNX export support for PaddleOCR-VL-1.5 and PP-Chart2Table architecture

#2432 Patch release to support transformers >= 5.0.0

#2427 Using from optimum.onnxruntime import ORTModelForSequenceClassification leads to ImportError: cannot import name 'is_tf_available' from 'transformers.utils'

#2423 Can't convert local timm models

#2417 Cannot run sub-commands on RHL because `optimum` mishandles the lib64 symlink to lib