All issues

Search results

#1 performance requirement

Status: Open.

#83In cmp-nct/ggllm.cpp;

· cmp-nct opened on Sep 27, 2023

Add support for AMX instructions (bf16 and/or int8)

Status: Open.

#81In cmp-nct/ggllm.cpp;

· WilliamTambellini opened on Aug 3, 2023

Metal support

Status: Open.

#78In cmp-nct/ggllm.cpp;

· ghost opened on Aug 2, 2023

Log the version of cuda that is being used

Status: Open.

#77In cmp-nct/ggllm.cpp;

· WilliamTambellini opened on Aug 1, 2023

linking error with static build

Status: Open.

#76In cmp-nct/ggllm.cpp;

· WilliamTambellini opened on Jul 31, 2023

Performance - heads up

enhancementNew feature or requestNew feature or request

Status: Open.

#75In cmp-nct/ggllm.cpp;

· cmp-nct opened on Jul 27, 2023

Would ggllm.cpp adopt the upcoming ggml fileformat (GGUF)?

Status: Open.

#72In cmp-nct/ggllm.cpp;

· klosax opened on Jul 21, 2023

--help , pipes and inconsistent help text

finished

Status: Open.

#71In cmp-nct/ggllm.cpp;

· maddes8cht opened on Jul 19, 2023

Can't falcon_convert on OpenBuddy Falcon 7B model, KeyError [fixed]

finished

good first issueGood for newcomersGood for newcomers

Status: Open.

#68In cmp-nct/ggllm.cpp;

· tak2hu opened on Jul 18, 2023

Even in interactive mode, multiturn conversation is not possible.

bugSomething isn't workingSomething isn't working

finished

Status: Open.

#67In cmp-nct/ggllm.cpp;

· ehalit opened on Jul 17, 2023

Could we also get the Makefile updated to build a libfalcon.so

Status: Open.

#64In cmp-nct/ggllm.cpp;

· linuxmagic-mp opened on Jul 14, 2023

Upcoming PR - Pushing the Context limit to 8k+ for all existing Falcon models - Longrange Falcon flights

enhancementNew feature or requestNew feature or request

finished

Status: Open.

#62In cmp-nct/ggllm.cpp;

· cmp-nct opened on Jul 13, 2023