DeepSeek·released 2025-01-201 source
Not yet benchmarked
Claims drawn from cited facts, not live model generation.
Developed by DeepSeek, this model was released in 2025. It contains 14.77 billion parameters, a mid-size scale that balances efficiency and lower cost with the performance headroom and expense of larger systems.
3 cited facts
This model is not yet benchmarked on any tracked evaluation, so it holds no current top score.
2 cited facts
It holds the top score on 0 benchmarks, meaning it has no category leadership yet. Its average gap to the SOTA leader is unknown, so no frontier/competitive/behind band is assigned.
2 cited facts
2 facts cross-checked across data sources: 2 single-source.
Source facts, citations, and refresh stamp for this record.
Sources: huggingface_models
DeepSeek-R1-Distill-Qwen-14B is an AI model developed by DeepSeek, released 2025-01-20. Its benchmark record below tags every score as independently reproduced, self-reported, or unverified.