turboderp 07ffea7f89 compare_q.py: Fix llama.cpp bpw measurement for MoE models

2025-05-18 00:19:59 +02:00

..

eval_texts

Add simple long-context evaluation script

2025-05-17 16:58:12 +02:00

spec

compare_q.py: Add more GPTQ layer types

2025-05-18 00:19:19 +02:00

compare_q_exllamav2.py

compare_q.py: Account for unquantized weights in blocksparse EXL2 layers

2025-05-14 23:55:25 +02:00

compare_q_exllamav3.py

compare_q.py: Add KLD test and some other tweaks

2025-05-16 16:13:26 +02:00

compare_q_llamacpp.py

compare_q.py: Fix llama.cpp bpw measurement for MoE models

2025-05-18 00:19:59 +02:00

compare_q_transformers.py

compare_q.py: Add more GPTQ layer types

2025-05-18 00:19:19 +02:00

compare_q.py

compare_q.py: Add KLD test and some other tweaks

2025-05-16 16:13:26 +02:00

humaneval.py

HumanEval: Move BOS token to individual prompt template, don't prepend by default when tokenizing

2025-05-11 23:02:07 +02:00

longctx.py

Add simple long-context evaluation script

2025-05-17 16:58:12 +02:00

model_diff.py

model_diff.py: Use deferred load and close file handles between modules

2025-05-12 21:23:48 +02:00

ppl.py

Reduce VRAM overhead in ppl test

2025-04-17 22:22:47 +02:00