Issues
- 2
Metal support
#78 opened by jempabroni - 7
- 0
#1 performance requirement
#83 opened by cmp-nct - 5
Upcoming PR - Pushing the Context limit to 8k+ for all existing Falcon models - Longrange Falcon flights
#62 opened by cmp-nct - 2
Performance - heads up
#75 opened by cmp-nct - 7
--help , pipes and inconsistent help text
#71 opened by maddes8cht - 2
linking error with static build
#76 opened by WilliamTambellini - 31
Unable to run TheBloke Falcon40b-instruct
#1 opened by iHaagcom - 0
- 0
- 13
Is there any GUI or Web UI for ggllm.cpp?
#52 opened by JohnClaw - 1
- 3
- 1
- 2
Debug Timings No Longer Working
#55 opened by boricuapab - 5
Instruct Mode Issue
#54 opened by boricuapab - 3
- 4
K Quant 64 support - quite a feat to integrate
#34 opened by cmp-nct - 4
Performance at high context (18k+)
#56 opened by cmp-nct - 4
OpenBLAS and CLBlast support
#41 opened by Fr0d0Beutl1n - 19
- 3
- 2
- 14
Steps forward - Tokenizer
#37 opened by cmp-nct - 12
Something weird is going on with -ngl
#12 opened by KerfuffleV2 - 6
Apple Silicon Unable To Build
#49 opened by only-cliches - 12
Mul_mat Speedup??
#31 opened by boricuapab - 0
Implement ChatGLM.cpp
#47 opened by iHaagcom - 4
Random spikes of up to 30ms in ggml_cuda_op device synchronization when using a low -ngl count with dual GPU
#19 opened by cmp-nct - 3
- 12
Problem with cMake on Linux focal, Cuda
#22 opened by linuxmagic-mp - 4
A strange delay happens after about 200 tokens
#15 opened by cmp-nct - 27
Slowdown with tokens
#6 opened by cmp-nct - 2
- 1
Windows Installation Video Tutorial
#29 opened by boricuapab - 2
CUDA mul_mat using cuBLAS for 3d multiplication fails on lm_head only for Falcon 7B
#30 opened by cmp-nct - 1
Parameter --reverse-prompt won't accept text
#24 opened by BorisEagle - 6
slow on 3090 and very high cpu usage
#17 opened by stupiding - 13
Illegal instruction (core dumped) on Ubuntu 22.04.2
#11 opened by dRAT3 - 1
mmap fails on WSL (linux too?)
#13 opened by cmp-nct - 2
Unable to make falcon_main
#10 opened by tomBlueOrange - 2
Why divert from the default GGML versions?
#5 opened by LLukas22 - 16