turboderp-org/exllamav3
1
0
Fork 0
mirror of https://github.com/turboderp-org/exllamav3.git synced 2026-07-12 02:05:10 +00:00
Code Issues Packages Projects Releases Wiki Activity
1,186 Commits 3 Branches 43 Tags
dev
Commit Graph

10 Commits

Author SHA1 Message Date
turboderp
6faf89f292 convert_mtp.py: Use mul1 codebook by default for standalone MTP convert 2026-07-03 00:08:37 +02:00
turboderp
9c5009efaa Add utility to convert MTP model separately (for legacy quants) 2026-06-09 22:15:21 +02:00
turboderp
f1edb0cff9 Util: Add template formatting test util 2026-06-05 03:16:47 +02:00
turboderp
a86b8f3ee9 Make tensor name patches a per-architecture option 2026-05-31 17:17:04 +02:00
Mamy Ratsimbazafy
e7ccf0fe2b fix regression: 'final_bits' in recompile.py - mirrors 2e52f80480 2026-04-15 21:34:04 +02:00
Mamy Ratsimbazafy
ad1aab406f Add size estimation script for model tensors size 2025-11-13 13:49:39 +01:00
turboderp
e7e06599fb Optimizer: Scripts 2025-09-11 22:16:02 +02:00
turboderp
6f77a33415 Add recompile utility script 2025-07-08 19:51:01 +02:00
turboderp
f6e2de2a7c Add utility to build model_safetensors_index.json for older quants that don't have one by default 2025-06-07 05:49:15 +02:00
turboderp
b73e159bc8 Add quantization_config.json 2025-04-08 15:27:17 +02:00