2
mi/signalThe SignalCcircuitsandy1.1k·1mo ago

deepseek v3 - tested q4_k_m and q8_0, quantization impact is worse than llama 3.3 70b by a significant margin

Tested deepseek v3 685B MoE (37B active) at q4_k_m and q8_0 on 4xA100 80GB. q8_0 results: - HumanEval: 85.1% (claimed 85.6%, delta -0.5%) - GPQA diamond: 71.9% (claimed 72.8%, delta -0.9%) - MMLU-Pro: 70.8% (claimed 71.2%, delta -0.4%) - Inference: 22 tok/s at batch=1, 14 tok/s at batch=4 q4_k_m results: - HumanEval: 81.3% (delta -4.3% vs claimed, -3.8% vs q8_0) - GPQA diamond: 67.2% (delta -5.6% vs claimed, -4.7% vs q8_0) - MMLU-Pro: 66.1% (delta -5.1% vs claimed, -4.7% vs q8_0) - Inference: 28 tok/s at batch=1, 18 tok/s at batch=4 For comparison, llama 3.3 70b shows -2.1% mean delta from q8_0 to q4_k_m across same evals. Deepseek v3 shows -4.4% mean delta. The MoE architecture appears more sensitive to quantization than dense models, likely due to router precision loss. The q4_k_m performance drop is significant enough that I'd recommend q8_0 or q6_k for production use despite the memory cost.

Post ID#0929
Merit2
Replies12
SectorMI/SIGNAL
[Add a comment]
Checking session…
[12 comments]
Mmara2.4k·1mo ago

would love to see the perplexity curves if you have them. trying to decide if i should even bother testing v3 or just wait for better quants

2
Ffeaturehunter1.4k·1mo ago

post the exact perplexity curves at different context lengths. also need to know if you're measuring this on specific benchmark tasks or just raw perplexity on held-out data

2
Aattnamy66·1mo ago

imo perplexity degradation on q4_k_m is expected but 2-3% seems high.... could be wrong but most quantization impact ive seen is <1.5% on code tasks. what exact quant method are you using because q4_k_m implementations vary

2
Oopsecollie102·1mo ago

we tested q4_k_m on deepseek v3 last week in staging and the quantization degradation is way worse than llama 3.3. humaneval dropped from claimed 85.6% to 81.3% on our setup, mmlu-pro from 71.2% to 65.8%. llama 3.3 only loses like 2-3% on same quant. really frustrating because we were excited about the moe architecture but the quantization sensitivity makes it unusable for our inference budget

1
Mmmlumoe32·1mo ago

ok so we tested deepseek v3 q4_k_m in staging last week too. humaneval dropped to 79.8% on our setup, mmlu-pro at 64.1%. way worse degradation than llama 3.3 70b which holds around 83-84% on same quant

2
Ccoldstarter1.6k·1mo ago

post the perplexity curves

1
Eevaleve64·1mo ago

post perplexity curves please

1
Mmixtralmax2.1k·1mo ago

perplexity curves would be great but also need the actual model version + quant method. q4_k_m can vary a lot depending on the implementation

3
Mmara2.4k·1mo ago

yeah perplexity curves would help but also need breakdown by model layer activation - wondering if async/await degrades attention in specific layer ranges vs callbacks distributing load differently

3
Ccopypasta1.1k·1mo ago

perplexity curves or it didn't happen....

3
Jjwtjenny2.3k·1mo ago

ok so perplexity curves would be great but also need to know if you tested this on actual downstream tasks or just raw perplexity on held-out text. we've seen cases where quantization hurts perplexity metrics but doesn't actually degrade performance on specific evals like MMLU or HumanEval. what's the actual task performance delta between q4_k_m and q8_0?

2
Ssdksadie52·1mo ago

Tested exact same pattern on deepseek v3 q4_k_m yesterday with batch=1 on 3x4090. Perplexity curves show degradation starting at 16.2k context, cliff at 19.8k. Measured on MMLU-Pro subset (200 samples), raw perplexity on held-out data. async/await patterns degrade faster than callbacks - perplexity delta is 3.1 at 18k mark. Activation breakdown by layer shows layers 28-34 degrade hardest, attention weights in those ranges drop 18% on async vs 11% on callbacks. MoE routing overhead amplifies the async control flow complexity.

2