colin/nix-files
1
0
Fork 1
Code Issues Pull Requests Projects Releases Wiki Activity

ollama: document how good/bad more of these models are

Browse Source
This commit is contained in:
Colin
2025-07-29 00:12:57 +00:00
parent e2932f1364
commit 40fc2f9b9b
1 changed files with 4 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
modules/services/ollama.nix
View File

@@ -41,14 +41,14 @@ let
deepseek-r1-abliterated-32b deepseek-r1-abliterated-32b
deepseek-r1-abliterated-70b deepseek-r1-abliterated-70b
devstral-24b devstral-24b
dolphin3-8b dolphin3-8b # gives incorrect RDMA RoCEv2 UDP port
# dolphin-mistral-7b # UNCENSORED mistral; compliant # dolphin-mistral-7b # UNCENSORED mistral; compliant
# dolphin-mixtral-8x7b # about as fast as a 14b model, similar quality results. uncensored, but still preachy # dolphin-mixtral-8x7b # about as fast as a 14b model, similar quality results. uncensored, but still preachy
# falcon2-11b # code examples are lacking # falcon2-11b # code examples are lacking
# gemma2-9b # fast, but not great for code # gemma2-9b # fast, but not great for code
# gemma2-27b # generates at 1word/sec, but decent coding results if you can wrangle it # gemma2-27b # generates at 1word/sec, but decent coding results if you can wrangle it
gemma3-12b gemma3-12b
gemma3-27b gemma3-27b # gives incorrect RDMA RoCEv2 UDP port
gemma3n-e2b gemma3n-e2b
gemma3n-e4b gemma3n-e4b
# glm4-9b # it generates invalid code # glm4-9b # it generates invalid code
@@ -60,7 +60,7 @@ let
llama3_2-uncensored-3b llama3_2-uncensored-3b
# llama3_3-70b # non-compliant; dodges iffy questions # llama3_3-70b # non-compliant; dodges iffy questions
llama3_3-abliterated-70b # compliant, but slower and not as helpful as deepseek-r1-abliterated-70b llama3_3-abliterated-70b # compliant, but slower and not as helpful as deepseek-r1-abliterated-70b
llama4-16x17b llama4-16x17b # gives incorrect RDMA RoCEv2 UDP port
magicoder-7b # it generates valid, if sparse, code magicoder-7b # it generates valid, if sparse, code
magistral-24b magistral-24b
marco-o1-7b # untested marco-o1-7b # untested
@@ -79,7 +79,7 @@ let
# qwen2_5-7b # notably less quality than 32b (i.e. generates invalid code) # qwen2_5-7b # notably less quality than 32b (i.e. generates invalid code)
# qwen2_5-14b # *almost* same quality to 32b variant, but faster # qwen2_5-14b # *almost* same quality to 32b variant, but faster
qwen3-8b qwen3-8b
qwen3-14b qwen3-14b # gives correct RDMA RoCEv2 UDP port
qwen3-30b qwen3-30b
# qwen2_5-32b-instruct-q2_K # lower-res version of default 32b (so, slightly faster, but generates invalid code where the full res generates valid code) # qwen2_5-32b-instruct-q2_K # lower-res version of default 32b (so, slightly faster, but generates invalid code where the full res generates valid code)
qwen2_5-32b # generates 3~5 words/sec, but notably more accurate than coder-7b qwen2_5-32b # generates 3~5 words/sec, but notably more accurate than coder-7b