Coding · DeepSeek
DeepSeek Coder V2 Lite
16B MoE (2.4B active) coder; quick on modest GPUs.
Datasheet
Maker | DeepSeek |
|---|---|
Parameters | 15.71 BVerifiedsafetensors.total from the HF API Source |
Licence | DeepSeek LicenseVerified |
Access | Public download |
Task (HF pipeline) | text-generation |
First published on HF | 2024-06-14 |
Last updated on HF | 2024-07-03 |
Weights in memory | 31 GB at 16-bit · 9.5 GB at ~4.85 bitsEstimateParameters × bits ÷ 8. Excludes context cache and runtime overhead; not every model has a 4-bit build. |
Benchmark
Not yet measured
SWE-bench Verified, % resolved
No comparable public score for this benchmark yet. We will not estimate one.
The description summarises the model card at https://huggingface.co/deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct. Parameters and licence are read from the Hugging Face API; the weights-in-memory line is our arithmetic, not a measurement.