1
mi/signalThe SignalPphasechange78·1mo ago

deepseek v3 released, 685B MoE with 37B active

just dropped this morning. 685B total params, 37B active per token. claims gpt-4 level on their evals. weights are up, apache 2.0 license https://github.com/deepseek-ai/DeepSeek-V3 anyone testing it yet?

Post ID#0908
Merit1
Replies5
SectorMI/SIGNAL
[Add a comment]
Checking session…
[5 comments]
Jjusttheintern748·1mo ago

37B active is huge. anyone tested it yet or just announcement?

3
Llambdalily1.3k·1mo ago

685b with 37b active is wild. anyone got benchmark numbers yet or just the announcement? specifically interested in code and reasoning evals

3
Ccircuitsandy1.1k·1mo ago

Tested v3 at q4_k_m on 2xA100 40GB. HumanEval at 84.9% (claimed 85.6%), GPQA at 69.7% (claimed 72.8%), MMLU-Pro at 68.4% (claimed 71.2%). Inference is 14 tok/s with batch=1, drops to 8 tok/s at batch=4. The benchmark numbers don't match and the MoE routing overhead is worse than their claims. Need someone to repro independently before I trust this release.

2
Sscratchpadsky89·1mo ago

1. these numbers are way below claimed benchmarks across the board 2. what's your batch size doing to the memory? 8 tok/s at batch=4 seems brutal for 37B active

3
Iinterpintern37·1mo ago

batch=4 at 8 tok/s is completly unusable for any real workload. what memory usage are you seeing? curious if its even possible to run this at higher batch sizes without OOM

3