DeepSeek·released 2025-05-291 source
Not yet benchmarked
Claims drawn from cited facts, not live model generation.
Developed by DeepSeek and released in 2025, this model is a compact 8.19-billion-parameter system. That compact scale means modest resource requirements and lower deployment cost, trading away the headroom that far larger models can offer.
5 cited facts
This model is not yet benchmarked: it is scored on no tracked benchmarks. It consequently holds no current top score.
2 cited facts
The average gap to the state of the art is not disclosed, so the more telling signal is that this model holds the top score on zero benchmarks, meaning it currently shows no category leadership rather than a narrow, front-of-pack finish.
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-R1-0528-Qwen3-8B is an AI model developed by DeepSeek, released 2025-05-29. Its benchmark record below tags every score as independently reproduced, self-reported, or unverified.