Skip to content

Pull requests: turboderp-org/exllamav2

New pull request New
21 Open 124 Closed
21 Open 124 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

DRY: Fix Greedy Clone typo in sampler.
#801 opened Jul 28, 2025 by Ph0rk0z
model load performance tweak idea
#799 opened Jul 10, 2025 by metaclassing
small tensor parallelism optimization
#796 opened Jun 14, 2025 by Alex-GZ
2
Hacky tensor parallel for Vision Models
#791 opened May 25, 2025 by Ph0rk0z • Draft
fix divide by zero error if there's no error in the optimization.
#783 opened Apr 30, 2025 by allenbenz
2
Fix ext.py silently ignoring ImportErrors
#774 opened Apr 24, 2025 by ThisIsPIRI
Llama-3_1-Nemotron 51B support
#726 opened Jan 28, 2025 by ymcki
12
Add more choices to quantization tool. Post processing after sim_anneal(). (optimizer.py/ext_quant.cpp)
#712 opened Jan 10, 2025 by Originalimoc
2
Implementation of logit threshold sampler and confidence breaker
#657 opened Oct 17, 2024 by anchortense
Add ExLlamaV2Sampler.Settings.logits_processor
#634 opened Sep 23, 2024 by lapp0
4
[kernel] switch from PYBIND11 to TORCH_LIBRARY
#617 opened Sep 6, 2024 by AlpinDale • Draft
added option for tokenized input to dynamic generator
#613 opened Sep 2, 2024 by KT313
Add batch processing to MMLU and Humaneval evaluation scripts to prevent OOM errors
#597 opened Aug 23, 2024 by LlamaEnjoyer
8
Adding stream to 1 kernel.
#590 opened Aug 14, 2024 by Narsil
4
Simple QuaRot proof of concept.
#407 opened Apr 11, 2024 by sgsdxzy
5
Refactor token healing initialization.
#330 opened Feb 10, 2024 by bjj
7
Update setup.py UX responses and module checks with helpful messaging
#277 opened Jan 15, 2024 by bgorlick
4
Repeat layers to create FrankenModels
#275 opened Jan 12, 2024 by dnhkng
35
add QuiP quant support
#217 opened Dec 7, 2023 by waters222
3
Adding return_lowest_perplexity
#206 opened Dec 3, 2023 by ziadloo
Add copilot server example
#23 opened Sep 13, 2023 by chenhunghan
9
ProTip! Add no:assignee to see everything that’s not assigned.