COMMITS
/ examples/models/llama/export_llama_lib.py May 15, 2025
J
Pipe in local_global attention (#10883)
Jack committed
J
Refactor quantize.py functions to remove args
Jack committed
May 14, 2025
J
Fix broken tests
Jack committed
P
[Executorch][llm] Enable local global attention in export_llama script (#10836)
pytorchbot committed
May 13, 2025
J
Refactor _get_source_transforms to remove args
Jack committed
J
Refactor _to_edge_and_lower_llama to remove args
Jack committed
April 29, 2025
J
Add Qwen3 0.6B, 1.7B, and 4B (#10539)
Jack committed
April 21, 2025
April 19, 2025
J
Remove args from LLMEdgeManager and misc cleanup
Jack committed
April 17, 2025
C
Instruct users to run llama for qnn to the active repro
cccclai committed
April 10, 2025
P
[Executorch][llama] Enable quantized sdpa (#10062)
pytorchbot committed
P
[Executorch][llama] Renamed quantized_kv_cache to custom_kv_cache (#10061)
pytorchbot committed
April 7, 2025
S
[ET-VK][ez] Allow logit linear layer to be lowered to Vulkan (#9951)
Sicheng Stephen Jia committed
April 3, 2025
C
forward fix
cccclai committed
April 2, 2025
S
Qualcomm AI Engine Direct - Add QNN support for to_edge_transform_and_lower (#9643)
shewu-quic committed
March 26, 2025
W
Qualcomm AI Engine Direct - Mimi Enablement Stage 1 (#9570)
winskuo-quic committed
J
Add buck target for hf_download
Jack committed
March 25, 2025
J
Download checkpoints from HuggingFace (#9538)
Jack committed
J
Rename phi-4-mini to phi_4_mini (#9539)
Jack committed
March 24, 2025
S
Bump torchao pin to enable dynamic shapes in lowbit (#9555)
Scott Roy committed
J
Add SmolLM (smollm2) (#9541)
Jack committed
S
Updates torchao pin to enable shared embedding quantization (#9548)
Scott Roy committed
March 22, 2025
J
Make export llama checkpoint and param optional
Jack committed
J
Fix xnnpack quantization discrepancy for non-fp32
Jack committed
March 21, 2025
J
Refactor dtype handling in export_llama
Jack committed
H
Fix pyre issues from non-mypy dirs
Hansong committed
March 19, 2025
M
Arm backend: Add initial Llama model test case (#8679)
Måns Nilsson committed
March 6, 2025
J
Fix phi4mini test model (#8971)
Jack committed
March 5, 2025
J
Add Phi-4-mini-instruct (#8856)
Jack committed
February 27, 2025
J
Use to_edge_lower_and_transform for XNNPack (#8624)
Jack committed
February 26, 2025
S
Revert #8501 and #8624 (#8716)
Scott Wolchok committed
February 25, 2025
J
Use to_edge_lower_and_transform for XNNPack (#8624)
Jack committed
February 6, 2025
L
Move print_delegate_info to devtools
Limin Tang committed
January 31, 2025
January 30, 2025
P
January 23, 2025
K
[ExecuTorch][BE] Split kv cache and SDPA for better code sharing
Kimish Patel committed
January 22, 2025
M
Print delegation info in export_llama in verbose (#7803)
Mengtao Yuan committed
January 16, 2025
K
[ExecuTorch][Llama] Split custom sdpa op and kv cache (#7412)
Kimish Patel committed
January 13, 2025
J
Make kv cache pos buffer name more specific (#7635)
Jack Zhang committed
January 10, 2025
J
Fix executorch kv cache incompatibility with to_executorch lowering (#7279)
Jack Zhang committed
January 7, 2025
D
Add support to export XNNPACK based static_llama
derekxu committed
December 16, 2024
S
Add compute unit to coreml llama export (#7280)
Scott Roy committed
December 13, 2024
M
Add torchao mps lowbit ops to llama runner (#7037)
Manuel Candales committed
December 2, 2024
P
Transform model to be able to use Attention Sink
pytorchbot committed
November 21, 2024
E
Fix type-safety of `torch.nn.Module` instances
Edward Z. Yang committed
November 14, 2024
J
Export TorchTune llama3_2_vision in ET (#5911)
Jack Zhang committed
P
[ET-VK][Llama] Apply XNNPACK partitoner as well when lowering to Vulkan (#6857)
pytorchbot committed
November 13, 2024
J
Accept model type parameter in export_llama
Jack Zhang committed
November 12, 2024
J
torch.export()-only export Llama arg (#6695)
Jack Zhang committed