-
Notifications
You must be signed in to change notification settings - Fork 262
Pull requests: huggingface/optimum-habana
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
[SW-224874] Reduce index_copy to fp8 in llama2 - QDQ flow
#2065
opened Jun 17, 2025 by
Tiefen-boop
Loading…
Skip unnecessary padding in text generation task
#2055
opened Jun 13, 2025 by
kyotoyx
Loading…
3 tasks
Extra Content: support FP8 model with bf16 KVCache do generation
#2052
opened Jun 12, 2025 by
astachowiczhabana
Loading…
Extra Content: add support for reduced model
#2050
opened Jun 12, 2025 by
astachowiczhabana
Loading…
Extra Content: enable accuracy benchmark using torch compile
#2049
opened Jun 12, 2025 by
astachowiczhabana
Loading…
Minor Documentation Updates and Comments Clarification
#2048
opened Jun 12, 2025 by
kilavvy
Loading…
Extra Content: fixed lost modules in regional compilation
#2047
opened Jun 12, 2025 by
astachowiczhabana
Loading…
Extra Content: enable_running_lm_eval_with_log_samples
#2046
opened Jun 12, 2025 by
astachowiczhabana
Loading…
Extra Content: Ifeval and MMLU now better supported
#2045
opened Jun 12, 2025 by
astachowiczhabana
Loading…
Extra Content: float inputs for Mixtral 8x7B
#2043
opened Jun 12, 2025 by
astachowiczhabana
Loading…
Extra Content: remove capture_pre_autograd_graph call
#2042
opened Jun 12, 2025 by
astachowiczhabana
Loading…
Revert "Extra Content: trust_remote_code True by default"
#2041
opened Jun 12, 2025 by
astachowiczhabana
Loading…
Revert "Extra Content: text-generation experimental dir"
#2040
opened Jun 12, 2025 by
astachowiczhabana
Loading…
Add Eval Script for LLaMA 3.2 and other Multimodal Models
#2037
opened Jun 12, 2025 by
jaygala223
Loading…
Fix PT_HPU_LAZY_MODE assertion to match updated default value
#2032
opened Jun 11, 2025 by
jasi306
Loading…
3 tasks
Refactor Qwen2 Family - FP32 SDPA and max_position_embedding
#2030
opened Jun 11, 2025 by
Wei-Lin-Intel
Loading…
3 tasks
Previous Next
ProTip!
Filter pull requests by the default branch with base:main.