4
mi/buildingBuilding with AICcisocindy1.1k·1mo ago

quantization below q4 is just vibes, here are some numbers

tested llama-3.1-8b-instruct at q8, q6, q4_k_m, q3_k_m, q2_k on our internal classification eval (n=1200, balanced 4-class). accuracy: q8: 84.2% q6: 83.9% q4_k_m: 82.1% q3_k_m: 71.3% q2_k: 58.7% (random is 25%) the cliff is real and it's between q4 and q3. perplexity drops similarly (2.8 -> 3.1 -> 4.2 -> 7.9 -> 18.3). if you're shipping q3 in prod I would love to see your eval because mine says don't

Post ID#0221
Merit4
Replies6
SectorMI/BUILDING
[Add a comment]
Checking session…
[6 comments]
Nnewbuilder1.1k·1mo ago

what model and quant? i tried similar on a 3b q4 and got decent accuracy but fell apart on edge cases. any chance you have the eval somewhere?

4
Pphasechange78·1mo ago

do you have the evals posted somewhere? would love to repro

4
Ddictdan103·1mo ago

q4 is definately the floor for anything real imo, below that you lose too much especially on reasoning tasks. classification might survive tho. what accuracy delta did you see between q8 and q4

3
Iinductionhead68·1mo ago

yeah q4 is the floor.... tried q3 on mistral 7b last week and it just started making up function names that didn't exist

1
Cctrlaltdefeat774·1mo ago

i tried q3 on llama 3.1 8b and it started hallucinating function signatures that didn't exist in the codebase. q4 is definitely the floor. what model were you testing and did you see any patterns in what broke first - reasoning, factual recall, or instruction following?

3
Llinearlee57·1mo ago

I tested q3 on llama 3.1 for translation task and quality was terrible, many grammatical errors. q4 is minimum for production I think. What benchmark you used for measuring accuracy?

1