Releases: tenstorrent/tt-forge-fe
Nightly 0.1.0.dev20250714
Install via pip
pip install tt_forge_fe==0.1.0.dev20250714 --pre --extra-index-url https://pypi.eng.aws.tenstorrent.com/
pip install tt_tvm==0.1.0.dev20250714 --pre --extra-index-url https://pypi.eng.aws.tenstorrent.com/
Docker container
ghcr.io/tenstorrent/tt-forge-fe-slim:0.1.0.dev20250714
More detailed instructions can be found in the Getting Started docker section
Test status: ✅
Workflow run: https://github.com/tenstorrent/tt-forge-fe/actions/runs/16225230338
- no changes
Files:
Model Compatibility
| frontend | model_name | Wormhole |
|---|---|---|
| tt-forge-fe | pt_vilt_dandelin_vilt_b32_finetuned_vqa_qa_hf | ✅ |
| tt-forge-fe | pt_albert_base_v1_token_cls_hf | ✅ |
| tt-forge-fe | pt_albert_textattack_albert_base_v2_imdb_seq_cls_hf | ✅ |
| tt-forge-fe | pt_bert_emrecan_bert_base_turkish_cased_mean_nli_stsb_tr_sentence_embed_gen_hf | ✅ |
| tt-forge-fe | pt_autoencoder_linear_img_enc_github | ✅ |
| tt-forge-fe | pt_efficientnet_efficientnet_b0_img_cls_timm | ✅ |
| tt-forge-fe | pt_ghostnet_ghostnet_100_img_cls_timm | ✅ |
| tt-forge-fe | pt_hrnet_hrnet_w18_small_v1_pose_estimation_osmr | ✅ |
| tt-forge-fe | pt_mobilenetv1_basic_img_cls_torchvision | ✅ |
| tt-forge-fe | pt_mobilenetv2_basic_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_resnet_50_img_cls_hf | ✅ |
| tt-forge-fe | pt_resnext_resnext101_32x8d_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_segformer_nvidia_mit_b0_img_cls_hf | ✅ |
| tt-forge-fe | pt_vit_google_vit_base_patch16_224_img_cls_hf | ✅ |
| tt-forge-fe | pt_vovnet_vovnet27s_img_cls_osmr | ✅ |
| tt-forge-fe | onnx_efficientnet_efficientnet_b0_img_cls_timm | ✅ |
| tt-forge-fe | onnx_mobilenetv2_mobilenetv2_100_img_cls_timm | ✅ |
| tt-forge-fe | pt_albert_base_v1_mlm_hf | ✅ |
| tt-forge-fe | pt_albert_base_v2_mlm_hf | ✅ |
| tt-forge-fe | pt_albert_base_v2_token_cls_hf | ✅ |
| tt-forge-fe | pt_bert_bert_base_uncased_mlm_hf | ✅ |
| tt-forge-fe | pt_bert_phiyodr_bert_large_finetuned_squad2_qa_hf | ✅ |
| tt-forge-fe | pt_distilbert_distilbert_base_cased_mlm_hf | ✅ |
| tt-forge-fe | pt_perceiverio_deepmind_language_perceiver_mlm_hf | ✅ |
| tt-forge-fe | pt_ghostnet_ghostnet_100_in1k_img_cls_timm | ✅ |
| tt-forge-fe | pt_mobilenetv3_mobilenet_v3_large_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_resnext_resnext50_32x4d_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_wideresnet_wide_resnet50_2_img_cls_torchvision | ✅ |
| tt-forge-fe | pt_xception_xception71_tf_in1k_img_cls_timm | ✅ |
| tt-forge-fe | onnx_resnet_50_img_cls_hf | ✅ |
| tt-forge-fe | onnx_segformer_nvidia_mit_b0_img_cls_hf | ✅ |
| tt-forge-fe | tf_resnet_resnet50_img_cls_keras | ✅ |
Nightly 0.1.0.dev20250713
Install via pip
pip install tt_forge_fe==0.1.0.dev20250713 --pre --extra-index-url https://pypi.eng.aws.tenstorrent.com/
pip install tt_tvm==0.1.0.dev20250713 --pre --extra-index-url https://pypi.eng.aws.tenstorrent.com/
Docker container
ghcr.io/tenstorrent/tt-forge-fe-slim:0.1.0.dev20250713
More detailed instructions can be found in the Getting Started docker section
Test status: ✅
Workflow run: https://github.com/tenstorrent/tt-forge-fe/actions/runs/16225230338
- no changes
Files:
Model Compatibility
| frontend | model_name | Wormhole |
|---|---|---|
| tt-forge-fe | pt_vilt_dandelin_vilt_b32_finetuned_vqa_qa_hf | ✅ |
| tt-forge-fe | pt_albert_base_v1_token_cls_hf | ✅ |
| tt-forge-fe | pt_albert_textattack_albert_base_v2_imdb_seq_cls_hf | ✅ |
| tt-forge-fe | pt_bert_emrecan_bert_base_turkish_cased_mean_nli_stsb_tr_sentence_embed_gen_hf | ✅ |
| tt-forge-fe | pt_autoencoder_linear_img_enc_github | ✅ |
| tt-forge-fe | pt_efficientnet_efficientnet_b0_img_cls_timm | ✅ |
| tt-forge-fe | pt_ghostnet_ghostnet_100_img_cls_timm | ✅ |
| tt-forge-fe | pt_hrnet_hrnet_w18_small_v1_pose_estimation_osmr | ✅ |
| tt-forge-fe | pt_mobilenetv1_basic_img_cls_torchvision | ✅ |
| tt-forge-fe | pt_mobilenetv2_basic_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_resnet_50_img_cls_hf | ✅ |
| tt-forge-fe | pt_resnext_resnext101_32x8d_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_segformer_nvidia_mit_b0_img_cls_hf | ✅ |
| tt-forge-fe | pt_vit_google_vit_base_patch16_224_img_cls_hf | ✅ |
| tt-forge-fe | pt_vovnet_vovnet27s_img_cls_osmr | ✅ |
| tt-forge-fe | onnx_efficientnet_efficientnet_b0_img_cls_timm | ✅ |
| tt-forge-fe | onnx_mobilenetv2_mobilenetv2_100_img_cls_timm | ✅ |
| tt-forge-fe | pt_albert_base_v1_mlm_hf | ✅ |
| tt-forge-fe | pt_albert_base_v2_mlm_hf | ✅ |
| tt-forge-fe | pt_albert_base_v2_token_cls_hf | ✅ |
| tt-forge-fe | pt_bert_bert_base_uncased_mlm_hf | ✅ |
| tt-forge-fe | pt_bert_phiyodr_bert_large_finetuned_squad2_qa_hf | ✅ |
| tt-forge-fe | pt_distilbert_distilbert_base_cased_mlm_hf | ✅ |
| tt-forge-fe | pt_perceiverio_deepmind_language_perceiver_mlm_hf | ✅ |
| tt-forge-fe | pt_ghostnet_ghostnet_100_in1k_img_cls_timm | ✅ |
| tt-forge-fe | pt_mobilenetv3_mobilenet_v3_large_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_resnext_resnext50_32x4d_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_wideresnet_wide_resnet50_2_img_cls_torchvision | ✅ |
| tt-forge-fe | pt_xception_xception71_tf_in1k_img_cls_timm | ✅ |
| tt-forge-fe | onnx_resnet_50_img_cls_hf | ✅ |
| tt-forge-fe | onnx_segformer_nvidia_mit_b0_img_cls_hf | ✅ |
| tt-forge-fe | tf_resnet_resnet50_img_cls_keras | ✅ |
Nightly 0.1.0.dev20250712
Install via pip
pip install tt_forge_fe==0.1.0.dev20250712 --pre --extra-index-url https://pypi.eng.aws.tenstorrent.com/
pip install tt_tvm==0.1.0.dev20250712 --pre --extra-index-url https://pypi.eng.aws.tenstorrent.com/
Docker container
ghcr.io/tenstorrent/tt-forge-fe-slim:0.1.0.dev20250712
More detailed instructions can be found in the Getting Started docker section
Test status: ✅
Workflow run: https://github.com/tenstorrent/tt-forge-fe/actions/runs/16225230338
🎯 Other Changes
- Remove model analysis docs generation pipeline (#2487) by @chandrasekaranpradeep
- [CI] Increase num test groups for BH nightly (#2497) by @nsumrakTT
- Fix onnx llama, phi and qwen models (#2493) by @chandrasekaranpradeep
- [OP] Divide migrated to cpp (#2482) by @vkovinicTT
- [cleanup] remove torch 2.0 integration (#2501) by @pilkicTT
- Run perf benchmark in tt-forge (#2452) by @nsumrakTT
- [deps] bump version of (#2499) by @pilkicTT
- [OP] Sine / Cosine migrated to cpp (#2514) by @mstojkovicTT
Files:
Model Compatibility
| frontend | model_name | Wormhole |
|---|---|---|
| tt-forge-fe | pt_vilt_dandelin_vilt_b32_finetuned_vqa_qa_hf | ✅ |
| tt-forge-fe | pt_albert_base_v1_token_cls_hf | ✅ |
| tt-forge-fe | pt_albert_textattack_albert_base_v2_imdb_seq_cls_hf | ✅ |
| tt-forge-fe | pt_bert_emrecan_bert_base_turkish_cased_mean_nli_stsb_tr_sentence_embed_gen_hf | ✅ |
| tt-forge-fe | pt_autoencoder_linear_img_enc_github | ✅ |
| tt-forge-fe | pt_efficientnet_efficientnet_b0_img_cls_timm | ✅ |
| tt-forge-fe | pt_ghostnet_ghostnet_100_img_cls_timm | ✅ |
| tt-forge-fe | pt_hrnet_hrnet_w18_small_v1_pose_estimation_osmr | ✅ |
| tt-forge-fe | pt_mobilenetv1_basic_img_cls_torchvision | ✅ |
| tt-forge-fe | pt_mobilenetv2_basic_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_resnet_50_img_cls_hf | ✅ |
| tt-forge-fe | pt_resnext_resnext101_32x8d_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_segformer_nvidia_mit_b0_img_cls_hf | ✅ |
| tt-forge-fe | pt_vit_google_vit_base_patch16_224_img_cls_hf | ✅ |
| tt-forge-fe | pt_vovnet_vovnet27s_img_cls_osmr | ✅ |
| tt-forge-fe | onnx_efficientnet_efficientnet_b0_img_cls_timm | ✅ |
| tt-forge-fe | onnx_mobilenetv2_mobilenetv2_100_img_cls_timm | ✅ |
| tt-forge-fe | pt_albert_base_v1_mlm_hf | ✅ |
| tt-forge-fe | pt_albert_base_v2_mlm_hf | ✅ |
| tt-forge-fe | pt_albert_base_v2_token_cls_hf | ✅ |
| tt-forge-fe | pt_bert_bert_base_uncased_mlm_hf | ✅ |
| tt-forge-fe | pt_bert_phiyodr_bert_large_finetuned_squad2_qa_hf | ✅ |
| tt-forge-fe | pt_distilbert_distilbert_base_cased_mlm_hf | ✅ |
| tt-forge-fe | pt_perceiverio_deepmind_language_perceiver_mlm_hf | ✅ |
| tt-forge-fe | pt_ghostnet_ghostnet_100_in1k_img_cls_timm | ✅ |
| tt-forge-fe | pt_mobilenetv3_mobilenet_v3_large_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_resnext_resnext50_32x4d_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_wideresnet_wide_resnet50_2_img_cls_torchvision | ✅ |
| tt-forge-fe | pt_xception_xception71_tf_in1k_img_cls_timm | ✅ |
| tt-forge-fe | onnx_resnet_50_img_cls_hf | ✅ |
| tt-forge-fe | onnx_segformer_nvidia_mit_b0_img_cls_hf | ✅ |
| tt-forge-fe | tf_resnet_resnet50_img_cls_keras | ✅ |
Nightly 0.1.0.dev20250711
Install via pip
pip install tt_forge_fe==0.1.0.dev20250711 --pre --extra-index-url https://pypi.eng.aws.tenstorrent.com/
pip install tt_tvm==0.1.0.dev20250711 --pre --extra-index-url https://pypi.eng.aws.tenstorrent.com/
Docker container
ghcr.io/tenstorrent/tt-forge-fe-slim:0.1.0.dev20250711
More detailed instructions can be found in the Getting Started docker section
Test status: ✅
Workflow run: https://github.com/tenstorrent/tt-forge-fe/actions/runs/16225230338
🎯 Other Changes
- Implemented reshape op in cpp. (#2481) by @acolicTT
- Remove xfail marker from stereo pytorch and address monodepth pcc issues (#2479) by @chandrasekaranpradeep
- Remove 'deprecated_verification' from the test plan (#2434) by @vobojevicTT
- Uplift third_party/tt-mlir to origin/main (42332a2b361f458595be7b91769a17b9a92b2fc7) 2025-07-10 (#2492) by @vmilosevic
- Simplified OpType attributes usage. (#2473) by @acolicTT
- Remove model analysis docs generation pipeline (#2487) by @chandrasekaranpradeep
- [CI] Increase num test groups for BH nightly (#2497) by @nsumrakTT
- Fix onnx llama, phi and qwen models (#2493) by @chandrasekaranpradeep
- [OP] Divide migrated to cpp (#2482) by @vkovinicTT
- [cleanup] remove torch 2.0 integration (#2501) by @pilkicTT
- Run perf benchmark in tt-forge (#2452) by @nsumrakTT
- [deps] bump version of (#2499) by @pilkicTT
- [OP] Sine / Cosine migrated to cpp (#2514) by @mstojkovicTT
Files:
Model Compatibility
| frontend | model_name | Wormhole |
|---|---|---|
| tt-forge-fe | pt_vilt_dandelin_vilt_b32_finetuned_vqa_qa_hf | ✅ |
| tt-forge-fe | pt_albert_base_v1_token_cls_hf | ✅ |
| tt-forge-fe | pt_albert_textattack_albert_base_v2_imdb_seq_cls_hf | ✅ |
| tt-forge-fe | pt_bert_emrecan_bert_base_turkish_cased_mean_nli_stsb_tr_sentence_embed_gen_hf | ✅ |
| tt-forge-fe | pt_autoencoder_linear_img_enc_github | ✅ |
| tt-forge-fe | pt_efficientnet_efficientnet_b0_img_cls_timm | ✅ |
| tt-forge-fe | pt_ghostnet_ghostnet_100_img_cls_timm | ✅ |
| tt-forge-fe | pt_hrnet_hrnet_w18_small_v1_pose_estimation_osmr | ✅ |
| tt-forge-fe | pt_mobilenetv1_basic_img_cls_torchvision | ✅ |
| tt-forge-fe | pt_mobilenetv2_basic_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_resnet_50_img_cls_hf | ✅ |
| tt-forge-fe | pt_resnext_resnext101_32x8d_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_segformer_nvidia_mit_b0_img_cls_hf | ✅ |
| tt-forge-fe | pt_vit_google_vit_base_patch16_224_img_cls_hf | ✅ |
| tt-forge-fe | pt_vovnet_vovnet27s_img_cls_osmr | ✅ |
| tt-forge-fe | onnx_efficientnet_efficientnet_b0_img_cls_timm | ✅ |
| tt-forge-fe | onnx_mobilenetv2_mobilenetv2_100_img_cls_timm | ✅ |
| tt-forge-fe | pt_albert_base_v1_mlm_hf | ✅ |
| tt-forge-fe | pt_albert_base_v2_mlm_hf | ✅ |
| tt-forge-fe | pt_albert_base_v2_token_cls_hf | ✅ |
| tt-forge-fe | pt_bert_bert_base_uncased_mlm_hf | ✅ |
| tt-forge-fe | pt_bert_phiyodr_bert_large_finetuned_squad2_qa_hf | ✅ |
| tt-forge-fe | pt_distilbert_distilbert_base_cased_mlm_hf | ✅ |
| tt-forge-fe | pt_perceiverio_deepmind_language_perceiver_mlm_hf | ✅ |
| tt-forge-fe | pt_ghostnet_ghostnet_100_in1k_img_cls_timm | ✅ |
| tt-forge-fe | pt_mobilenetv3_mobilenet_v3_large_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_resnext_resnext50_32x4d_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_wideresnet_wide_resnet50_2_img_cls_torchvision | ✅ |
| tt-forge-fe | pt_xception_xception71_tf_in1k_img_cls_timm | ✅ |
| tt-forge-fe | onnx_resnet_50_img_cls_hf | ✅ |
| tt-forge-fe | onnx_segformer_nvidia_mit_b0_img_cls_hf | ✅ |
| tt-forge-fe | tf_resnet_resnet50_img_cls_keras | ✅ |
Nightly 0.1.0.dev20250710
Install via pip
pip install tt_forge_fe==0.1.0.dev20250710 --pre --extra-index-url https://pypi.eng.aws.tenstorrent.com/
pip install tt_tvm==0.1.0.dev20250710 --pre --extra-index-url https://pypi.eng.aws.tenstorrent.com/
Docker container
ghcr.io/tenstorrent/tt-forge-fe-slim:0.1.0.dev20250710
More detailed instructions can be found in the Getting Started docker section
Test status: ✅
Workflow run: https://github.com/tenstorrent/tt-forge-fe/actions/runs/16170389294
🎯 Other Changes
- Uplift third_party/tt-mlir to origin/main (4122d1c6c6f202b188fccbe0a244afd1aad2fd74) 2025-07-09 (#2485) by @vmilosevic
- Implement subtract op in cpp (#2471) by @mstojkovicTT
- Fix transformers update failures (#2421) by @chandrasekaranpradeep
Files:
Model Compatibility
| frontend | model_name | Wormhole |
|---|---|---|
| tt-forge-fe | pt_vilt_dandelin_vilt_b32_finetuned_vqa_qa_hf | ✅ |
| tt-forge-fe | pt_albert_base_v1_token_cls_hf | ✅ |
| tt-forge-fe | pt_albert_textattack_albert_base_v2_imdb_seq_cls_hf | ✅ |
| tt-forge-fe | pt_bert_emrecan_bert_base_turkish_cased_mean_nli_stsb_tr_sentence_embed_gen_hf | ✅ |
| tt-forge-fe | pt_autoencoder_linear_img_enc_github | ✅ |
| tt-forge-fe | pt_efficientnet_efficientnet_b0_img_cls_timm | ✅ |
| tt-forge-fe | pt_ghostnet_ghostnet_100_img_cls_timm | ✅ |
| tt-forge-fe | pt_hrnet_hrnet_w18_small_v1_pose_estimation_osmr | ✅ |
| tt-forge-fe | pt_mobilenetv1_basic_img_cls_torchvision | ✅ |
| tt-forge-fe | pt_mobilenetv2_basic_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_resnet_50_img_cls_hf | ✅ |
| tt-forge-fe | pt_resnext_resnext101_32x8d_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_segformer_nvidia_mit_b0_img_cls_hf | ✅ |
| tt-forge-fe | pt_vit_google_vit_base_patch16_224_img_cls_hf | ✅ |
| tt-forge-fe | pt_vovnet_vovnet27s_img_cls_osmr | ✅ |
| tt-forge-fe | onnx_efficientnet_efficientnet_b0_img_cls_timm | ✅ |
| tt-forge-fe | onnx_mobilenetv2_mobilenetv2_100_img_cls_timm | ✅ |
| tt-forge-fe | pt_albert_base_v1_mlm_hf | ✅ |
| tt-forge-fe | pt_albert_base_v2_mlm_hf | ✅ |
| tt-forge-fe | pt_albert_base_v2_token_cls_hf | ✅ |
| tt-forge-fe | pt_bert_bert_base_uncased_mlm_hf | ✅ |
| tt-forge-fe | pt_bert_phiyodr_bert_large_finetuned_squad2_qa_hf | ✅ |
| tt-forge-fe | pt_distilbert_distilbert_base_cased_mlm_hf | ✅ |
| tt-forge-fe | pt_perceiverio_deepmind_language_perceiver_mlm_hf | ✅ |
| tt-forge-fe | pt_ghostnet_ghostnet_100_in1k_img_cls_timm | ✅ |
| tt-forge-fe | pt_mobilenetv3_mobilenet_v3_large_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_resnext_resnext50_32x4d_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_wideresnet_wide_resnet50_2_img_cls_torchvision | ✅ |
| tt-forge-fe | pt_xception_xception71_tf_in1k_img_cls_timm | ✅ |
| tt-forge-fe | onnx_resnet_50_img_cls_hf | ✅ |
| tt-forge-fe | onnx_segformer_nvidia_mit_b0_img_cls_hf | ✅ |
| tt-forge-fe | tf_resnet_resnet50_img_cls_keras | ✅ |
Nightly 0.1.0.dev20250709
Install via pip
pip install tt_forge_fe==0.1.0.dev20250709 --pre --extra-index-url https://pypi.eng.aws.tenstorrent.com/
pip install tt_tvm==0.1.0.dev20250709 --pre --extra-index-url https://pypi.eng.aws.tenstorrent.com/
Docker container
ghcr.io/tenstorrent/tt-forge-fe-slim:0.1.0.dev20250709
More detailed instructions can be found in the Getting Started docker section
Test status: ✅
Workflow run: https://github.com/tenstorrent/tt-forge-fe/actions/runs/16150402839
🎯 Other Changes
- Run perf benchmark optionally on PR (#2468) by @nsumrakTT
- Uplift third_party/tt-mlir to origin/main (a1985cad3746150b5e1708ea739b562019a0a2cf) 2025-07-08 (#2477) by @vmilosevic
- Remove xfail markers for passing nightly tests (#2464) by @chandrasekaranpradeep
- [OP] Add migrated to cpp (#2462) by @vkovinicTT
Files:
Model Compatibility
| frontend | model_name | Wormhole |
|---|---|---|
| tt-forge-fe | pt_vilt_dandelin_vilt_b32_finetuned_vqa_qa_hf | ✅ |
| tt-forge-fe | pt_albert_base_v1_token_cls_hf | ✅ |
| tt-forge-fe | pt_albert_textattack_albert_base_v2_imdb_seq_cls_hf | ✅ |
| tt-forge-fe | pt_bert_emrecan_bert_base_turkish_cased_mean_nli_stsb_tr_sentence_embed_gen_hf | ✅ |
| tt-forge-fe | pt_autoencoder_linear_img_enc_github | ✅ |
| tt-forge-fe | pt_efficientnet_efficientnet_b0_img_cls_timm | ✅ |
| tt-forge-fe | pt_ghostnet_ghostnet_100_img_cls_timm | ✅ |
| tt-forge-fe | pt_hrnet_hrnet_w18_small_v1_pose_estimation_osmr | ✅ |
| tt-forge-fe | pt_mobilenetv1_basic_img_cls_torchvision | ✅ |
| tt-forge-fe | pt_mobilenetv2_basic_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_resnet_50_img_cls_hf | ✅ |
| tt-forge-fe | pt_resnext_resnext101_32x8d_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_segformer_nvidia_mit_b0_img_cls_hf | ✅ |
| tt-forge-fe | pt_vit_google_vit_base_patch16_224_img_cls_hf | ✅ |
| tt-forge-fe | pt_vovnet_vovnet27s_img_cls_osmr | ✅ |
| tt-forge-fe | onnx_efficientnet_efficientnet_b0_img_cls_timm | ✅ |
| tt-forge-fe | onnx_mobilenetv2_mobilenetv2_100_img_cls_timm | ✅ |
| tt-forge-fe | pt_albert_base_v1_mlm_hf | ✅ |
| tt-forge-fe | pt_albert_base_v2_mlm_hf | ✅ |
| tt-forge-fe | pt_albert_base_v2_token_cls_hf | ✅ |
| tt-forge-fe | pt_bert_bert_base_uncased_mlm_hf | ✅ |
| tt-forge-fe | pt_bert_phiyodr_bert_large_finetuned_squad2_qa_hf | ✅ |
| tt-forge-fe | pt_distilbert_distilbert_base_cased_mlm_hf | ✅ |
| tt-forge-fe | pt_perceiverio_deepmind_language_perceiver_mlm_hf | ✅ |
| tt-forge-fe | pt_ghostnet_ghostnet_100_in1k_img_cls_timm | ✅ |
| tt-forge-fe | pt_mobilenetv3_mobilenet_v3_large_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_resnext_resnext50_32x4d_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_wideresnet_wide_resnet50_2_img_cls_torchvision | ✅ |
| tt-forge-fe | pt_xception_xception71_tf_in1k_img_cls_timm | ✅ |
| tt-forge-fe | onnx_resnet_50_img_cls_hf | ✅ |
| tt-forge-fe | onnx_segformer_nvidia_mit_b0_img_cls_hf | ✅ |
| tt-forge-fe | tf_resnet_resnet50_img_cls_keras | ✅ |
Nightly 0.1.0.dev20250708
Install via pip
pip install tt_forge_fe==0.1.0.dev20250708 --pre --extra-index-url https://pypi.eng.aws.tenstorrent.com/
pip install tt_tvm==0.1.0.dev20250708 --pre --extra-index-url https://pypi.eng.aws.tenstorrent.com/
Docker container
ghcr.io/tenstorrent/tt-forge-fe-slim:0.1.0.dev20250708
More detailed instructions can be found in the Getting Started docker section
Test status: ✅
Workflow run: https://github.com/tenstorrent/tt-forge-fe/actions/runs/16126826596
🎯 Other Changes
- Uplift third_party/tt-mlir to origin/main (107cac17757a708b85cbea8a064a4d45853064c5) 2025-07-07 (#2463) by @vmilosevic
- [ops] fix malformed index op (#2457) by @pilkicTT
- [test] fix failing unit test in release mode (#2458) by @pilkicTT
- [ci] nightly maintenance (#2459) by @pilkicTT
- Implement multiply op in cpp (#2422) by @mstojkovicTT
- Replaced string comparison to op type comparison for ops. (#2455) by @acolicTT
- [ops] migrate transpose to new op infra (#2460) by @pilkicTT
- Renamed op type members. (#2469) by @acolicTT
Files:
Model Compatibility
| frontend | model_name | Wormhole |
|---|---|---|
| tt-forge-fe | pt_vilt_dandelin_vilt_b32_finetuned_vqa_qa_hf | ✅ |
| tt-forge-fe | pt_albert_base_v1_token_cls_hf | ✅ |
| tt-forge-fe | pt_albert_textattack_albert_base_v2_imdb_seq_cls_hf | ✅ |
| tt-forge-fe | pt_bert_emrecan_bert_base_turkish_cased_mean_nli_stsb_tr_sentence_embed_gen_hf | ✅ |
| tt-forge-fe | pt_autoencoder_linear_img_enc_github | ✅ |
| tt-forge-fe | pt_efficientnet_efficientnet_b0_img_cls_timm | ✅ |
| tt-forge-fe | pt_ghostnet_ghostnet_100_img_cls_timm | ✅ |
| tt-forge-fe | pt_hrnet_hrnet_w18_small_v1_pose_estimation_osmr | ✅ |
| tt-forge-fe | pt_mobilenetv1_basic_img_cls_torchvision | ✅ |
| tt-forge-fe | pt_mobilenetv2_basic_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_resnet_50_img_cls_hf | ✅ |
| tt-forge-fe | pt_resnext_resnext101_32x8d_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_segformer_nvidia_mit_b0_img_cls_hf | ✅ |
| tt-forge-fe | pt_vit_google_vit_base_patch16_224_img_cls_hf | ✅ |
| tt-forge-fe | pt_vovnet_vovnet27s_img_cls_osmr | ✅ |
| tt-forge-fe | onnx_efficientnet_efficientnet_b0_img_cls_timm | ✅ |
| tt-forge-fe | onnx_mobilenetv2_mobilenetv2_100_img_cls_timm | ✅ |
| tt-forge-fe | pt_albert_base_v1_mlm_hf | ✅ |
| tt-forge-fe | pt_albert_base_v2_mlm_hf | ✅ |
| tt-forge-fe | pt_albert_base_v2_token_cls_hf | ✅ |
| tt-forge-fe | pt_bert_bert_base_uncased_mlm_hf | ✅ |
| tt-forge-fe | pt_bert_phiyodr_bert_large_finetuned_squad2_qa_hf | ✅ |
| tt-forge-fe | pt_distilbert_distilbert_base_cased_mlm_hf | ✅ |
| tt-forge-fe | pt_perceiverio_deepmind_language_perceiver_mlm_hf | ✅ |
| tt-forge-fe | pt_ghostnet_ghostnet_100_in1k_img_cls_timm | ✅ |
| tt-forge-fe | pt_mobilenetv3_mobilenet_v3_large_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_resnext_resnext50_32x4d_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_wideresnet_wide_resnet50_2_img_cls_torchvision | ✅ |
| tt-forge-fe | pt_xception_xception71_tf_in1k_img_cls_timm | ✅ |
| tt-forge-fe | onnx_resnet_50_img_cls_hf | ✅ |
| tt-forge-fe | onnx_segformer_nvidia_mit_b0_img_cls_hf | ✅ |
| tt-forge-fe | tf_resnet_resnet50_img_cls_keras | ✅ |
Nightly 0.1.0.dev20250707
Install via pip
pip install tt_forge_fe==0.1.0.dev20250707 --pre --extra-index-url https://pypi.eng.aws.tenstorrent.com/
pip install tt_tvm==0.1.0.dev20250707 --pre --extra-index-url https://pypi.eng.aws.tenstorrent.com/
Docker container
ghcr.io/tenstorrent/tt-forge-fe-slim:0.1.0.dev20250707
More detailed instructions can be found in the Getting Started docker section
Test status: ✅
Workflow run: https://github.com/tenstorrent/tt-forge-fe/actions/runs/16097652760
🎯 Other Changes
- Uplift third_party/tt-mlir to origin/main (7e54f7fc197c05836602a929622de93df0e2f5a8) 2025-07-06 (#2461) by @vmilosevic
Files:
Model Compatibility
| frontend | model_name | Wormhole |
|---|---|---|
| tt-forge-fe | pt_albert_base_v1_mlm_hf | ✅ |
| tt-forge-fe | pt_albert_base_v2_mlm_hf | ✅ |
| tt-forge-fe | pt_albert_base_v2_token_cls_hf | ✅ |
| tt-forge-fe | pt_bert_bert_base_uncased_mlm_hf | ✅ |
| tt-forge-fe | pt_bert_phiyodr_bert_large_finetuned_squad2_qa_hf | ✅ |
| tt-forge-fe | pt_distilbert_distilbert_base_cased_mlm_hf | ✅ |
| tt-forge-fe | pt_perceiverio_deepmind_language_perceiver_mlm_hf | ✅ |
| tt-forge-fe | pt_ghostnet_ghostnet_100_in1k_img_cls_timm | ✅ |
| tt-forge-fe | pt_mobilenetv3_mobilenet_v3_large_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_resnext_resnext50_32x4d_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_wideresnet_wide_resnet50_2_img_cls_torchvision | ✅ |
| tt-forge-fe | pt_xception_xception71_tf_in1k_img_cls_timm | ✅ |
| tt-forge-fe | onnx_resnet_50_img_cls_hf | ✅ |
| tt-forge-fe | onnx_segformer_nvidia_mit_b0_img_cls_hf | ✅ |
| tt-forge-fe | tf_resnet_resnet50_img_cls_keras | ✅ |
| tt-forge-fe | pt_vilt_dandelin_vilt_b32_finetuned_vqa_qa_hf | ✅ |
| tt-forge-fe | pt_albert_base_v1_token_cls_hf | ✅ |
| tt-forge-fe | pt_albert_textattack_albert_base_v2_imdb_seq_cls_hf | ✅ |
| tt-forge-fe | pt_bert_emrecan_bert_base_turkish_cased_mean_nli_stsb_tr_sentence_embed_gen_hf | ✅ |
| tt-forge-fe | pt_autoencoder_linear_img_enc_github | ✅ |
| tt-forge-fe | pt_efficientnet_efficientnet_b0_img_cls_timm | ✅ |
| tt-forge-fe | pt_ghostnet_ghostnet_100_img_cls_timm | ✅ |
| tt-forge-fe | pt_hrnet_hrnet_w18_small_v1_pose_estimation_osmr | ✅ |
| tt-forge-fe | pt_mobilenetv1_basic_img_cls_torchvision | ✅ |
| tt-forge-fe | pt_mobilenetv2_basic_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_resnet_50_img_cls_hf | ✅ |
| tt-forge-fe | pt_resnext_resnext101_32x8d_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_segformer_nvidia_mit_b0_img_cls_hf | ✅ |
| tt-forge-fe | pt_vit_google_vit_base_patch16_224_img_cls_hf | ✅ |
| tt-forge-fe | pt_vovnet_vovnet27s_img_cls_osmr | ✅ |
| tt-forge-fe | onnx_efficientnet_efficientnet_b0_img_cls_timm | ✅ |
| tt-forge-fe | onnx_mobilenetv2_mobilenetv2_100_img_cls_timm | ✅ |
Nightly 0.1.0.dev20250706
Install via pip
pip install tt_forge_fe==0.1.0.dev20250706 --pre --extra-index-url https://pypi.eng.aws.tenstorrent.com/
pip install tt_tvm==0.1.0.dev20250706 --pre --extra-index-url https://pypi.eng.aws.tenstorrent.com/
Docker container
ghcr.io/tenstorrent/tt-forge-fe-slim:0.1.0.dev20250706
More detailed instructions can be found in the Getting Started docker section
Test status: ✅
Workflow run: https://github.com/tenstorrent/tt-forge-fe/actions/runs/16086984515
🎯 Other Changes
- Uplift third_party/tt-mlir to origin/main (ef912af9105139ee84f946bd1d9fe17f511baddb) 2025-07-05 (#2456) by @vmilosevic
Files:
Model Compatibility
| frontend | model_name | Wormhole |
|---|---|---|
| tt-forge-fe | pt_vilt_dandelin_vilt_b32_finetuned_vqa_qa_hf | ✅ |
| tt-forge-fe | pt_albert_base_v1_token_cls_hf | ✅ |
| tt-forge-fe | pt_albert_textattack_albert_base_v2_imdb_seq_cls_hf | ✅ |
| tt-forge-fe | pt_bert_emrecan_bert_base_turkish_cased_mean_nli_stsb_tr_sentence_embed_gen_hf | ✅ |
| tt-forge-fe | pt_autoencoder_linear_img_enc_github | ✅ |
| tt-forge-fe | pt_efficientnet_efficientnet_b0_img_cls_timm | ✅ |
| tt-forge-fe | pt_ghostnet_ghostnet_100_img_cls_timm | ✅ |
| tt-forge-fe | pt_hrnet_hrnet_w18_small_v1_pose_estimation_osmr | ✅ |
| tt-forge-fe | pt_mobilenetv1_basic_img_cls_torchvision | ✅ |
| tt-forge-fe | pt_mobilenetv2_basic_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_resnet_50_img_cls_hf | ✅ |
| tt-forge-fe | pt_resnext_resnext101_32x8d_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_segformer_nvidia_mit_b0_img_cls_hf | ✅ |
| tt-forge-fe | pt_vit_google_vit_base_patch16_224_img_cls_hf | ✅ |
| tt-forge-fe | pt_vovnet_vovnet27s_img_cls_osmr | ✅ |
| tt-forge-fe | onnx_efficientnet_efficientnet_b0_img_cls_timm | ✅ |
| tt-forge-fe | onnx_mobilenetv2_mobilenetv2_100_img_cls_timm | ✅ |
| tt-forge-fe | pt_albert_base_v1_mlm_hf | ✅ |
| tt-forge-fe | pt_albert_base_v2_mlm_hf | ✅ |
| tt-forge-fe | pt_albert_base_v2_token_cls_hf | ✅ |
| tt-forge-fe | pt_bert_bert_base_uncased_mlm_hf | ✅ |
| tt-forge-fe | pt_bert_phiyodr_bert_large_finetuned_squad2_qa_hf | ✅ |
| tt-forge-fe | pt_distilbert_distilbert_base_cased_mlm_hf | ✅ |
| tt-forge-fe | pt_perceiverio_deepmind_language_perceiver_mlm_hf | ✅ |
| tt-forge-fe | pt_ghostnet_ghostnet_100_in1k_img_cls_timm | ✅ |
| tt-forge-fe | pt_mobilenetv3_mobilenet_v3_large_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_resnext_resnext50_32x4d_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_wideresnet_wide_resnet50_2_img_cls_torchvision | ✅ |
| tt-forge-fe | pt_xception_xception71_tf_in1k_img_cls_timm | ✅ |
| tt-forge-fe | onnx_resnet_50_img_cls_hf | ✅ |
| tt-forge-fe | onnx_segformer_nvidia_mit_b0_img_cls_hf | ✅ |
| tt-forge-fe | tf_resnet_resnet50_img_cls_keras | ✅ |
Nightly 0.1.0.dev20250705
Install via pip
pip install tt_forge_fe==0.1.0.dev20250705 --pre --extra-index-url https://pypi.eng.aws.tenstorrent.com/
pip install tt_tvm==0.1.0.dev20250705 --pre --extra-index-url https://pypi.eng.aws.tenstorrent.com/
Docker container
ghcr.io/tenstorrent/tt-forge-fe-slim:0.1.0.dev20250705
More detailed instructions can be found in the Getting Started docker section
Test status: ❌
Workflow run: https://github.com/tenstorrent/tt-forge-fe/actions/runs/16080219682
🎯 Other Changes
- [Bug Fix] Fix decompose_nd_reshape_split pass (#2413) by @vkovinicTT
- add list of tt-mlir commits between uplifts (#2415) by @nsumrakTT
- [watchdog] show timeout exceeded message (#2449) by @pilkicTT
- Skip crashing sweeps tests (#1992) by @vbrkicTT
- Implemented missing decomposing functions in cpp ops. (#2451) by @acolicTT
- Uplift third_party/tt-mlir to origin/main 2025-07-04 (#2425) by @vmilosevic
- Update perf benchmark tests (#2432) by @rpavlovicTT
Files:
Model Compatibility
| frontend | model_name | Wormhole |
|---|---|---|
| tt-forge-fe | pt_vilt_dandelin_vilt_b32_finetuned_vqa_qa_hf | ✅ |
| tt-forge-fe | pt_albert_base_v1_token_cls_hf | ✅ |
| tt-forge-fe | pt_albert_textattack_albert_base_v2_imdb_seq_cls_hf | ✅ |
| tt-forge-fe | pt_bert_emrecan_bert_base_turkish_cased_mean_nli_stsb_tr_sentence_embed_gen_hf | ✅ |
| tt-forge-fe | pt_autoencoder_linear_img_enc_github | ✅ |
| tt-forge-fe | pt_efficientnet_efficientnet_b0_img_cls_timm | ✅ |
| tt-forge-fe | pt_ghostnet_ghostnet_100_img_cls_timm | ✅ |
| tt-forge-fe | pt_hrnet_hrnet_w18_small_v1_pose_estimation_osmr | ✅ |
| tt-forge-fe | pt_mobilenetv1_basic_img_cls_torchvision | ✅ |
| tt-forge-fe | pt_mobilenetv2_basic_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_resnet_50_img_cls_hf | ✅ |
| tt-forge-fe | pt_resnext_resnext101_32x8d_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_segformer_nvidia_mit_b0_img_cls_hf | ✅ |
| tt-forge-fe | pt_vit_google_vit_base_patch16_224_img_cls_hf | ✅ |
| tt-forge-fe | pt_vovnet_vovnet27s_img_cls_osmr | ✅ |
| tt-forge-fe | onnx_efficientnet_efficientnet_b0_img_cls_timm | ✅ |
| tt-forge-fe | onnx_mobilenetv2_mobilenetv2_100_img_cls_timm | ✅ |
| tt-forge-fe | pt_albert_base_v1_mlm_hf | ✅ |
| tt-forge-fe | pt_albert_base_v2_mlm_hf | ✅ |
| tt-forge-fe | pt_albert_base_v2_token_cls_hf | ✅ |
| tt-forge-fe | pt_bert_bert_base_uncased_mlm_hf | ✅ |
| tt-forge-fe | pt_bert_phiyodr_bert_large_finetuned_squad2_qa_hf | ✅ |
| tt-forge-fe | pt_distilbert_distilbert_base_cased_mlm_hf | ✅ |
| tt-forge-fe | pt_perceiverio_deepmind_language_perceiver_mlm_hf | ✅ |
| tt-forge-fe | pt_ghostnet_ghostnet_100_in1k_img_cls_timm | ✅ |
| tt-forge-fe | pt_mobilenetv3_mobilenet_v3_large_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_resnext_resnext50_32x4d_img_cls_torchhub | ✅ |
| tt-forge-fe | pt_wideresnet_wide_resnet50_2_img_cls_torchvision | ✅ |
| tt-forge-fe | pt_xception_xception71_tf_in1k_img_cls_timm | ✅ |
| tt-forge-fe | onnx_resnet_50_img_cls_hf | ✅ |
| tt-forge-fe | onnx_segformer_nvidia_mit_b0_img_cls_hf | ✅ |
| tt-forge-fe | tf_resnet_resnet50_img_cls_keras | ✅ |