@ashfold: Revealing the answer. While running the dim-agent benchmark, we noticed that DSv4's scores have been consistently improving. The whales are cooking!
Summary
While running the dim-agent benchmark, the author noticed that DSv4's scores have been consistently improving, hinting at significant progress in model development.
View Cached Full Text
Cached at: 06/23/26, 08:03 AM
Reveal the answer. While benchmarking dim-agent, we noticed that DSv4’s performance has been steadily improving. The whales are cooking! https://t.co/RXsnNAcEXv
Similar Articles
A few words on DS4
Antirez announces DwarfStar 4 (DS4), a local AI tool that runs DeepSeek v4 Flash with asymmetric 2/8 bit quantization on high-end consumer hardware, achieving near-frontier performance. He discusses the project's rapid popularity, future plans for model updates and distributed inference, and the significance of local AI for serious tasks.
@ArizePhoenix: The result: a 50% improvement over 5.2 on their in-house Code Bench, Terminal-Bench 3.0 from 4.6 → 28.3, DeepSWE v1.1 f…
DeepSWE v1.1 shows a 50% improvement on in-house Code Bench and sets new state-of-the-art scores on Terminal-Bench 3.0 and Agents' Last Exam, with emerging cyber capabilities advancing faster than expected.
@antirez: DS4 running on DGX Spark (GB10 / CUDA), private branch for now. 12 tokens/sec, the memory bandwidth is limited in this …
Antirez reports benchmarking DS4 inference on the DGX Spark (GB10), noting 12 tokens/sec generation speed and high prefill performance, with plans to merge the codebase once mature.
@danielhanchen: DeepSeek just released DSpark for V4 Flash & Pro, a new speculative decoding method boosting throughput by 51% to 400%!…
DeepSeek released DSpark, a speculative decoding method that boosts throughput by 51% to 400% for V4 Flash & Pro, along with the open-source DeepSpec codebase for training and evaluating draft models.
@danieltvela: This model is going to be amazing.
DeepSeek-V4-Flash official API is now live in public beta, featuring massively upgraded agent capabilities and benchmark scores surpassing V4-Pro-Preview.