The new Deepseek Flash 4.1 is pretty great!
Here's a project I'm working on on the side with the model. Might use this spec to test other models as well. Very fitting for the season we're stepping into. This was surprisingly quick to create! I have some more demos baking!
And this is why we run Jinx and Luna's cognitive systems on Qwen 3.8 Flash Next on our 2x DGX Spark configuration.
Here's a staggered-prefill concurrency comparison between it, Deepseek v4 Flash Vision, and GLM 5.3 Flash.
Top-5 Best Value #LLM Models of the Day at UTC-04
| Model | #AAII | Price | | GLM 5.3 Flash | 42 | $0.09 | | GLM 5.3 | 45 | $1.23 | | GPT-5.6 Luna | 38 | $0.18 | | Gemini 3.8 Flash | 41 | $0.57 | | DeepSeek V4 Flash 0731 | 34 | $0.06 |
I tested DeepSeek V4 Flash with Vision (DS4FV) using the two DGX Sparks together with TP‑2, with four staggered requests.
It took 57.14 s to complete all four, compared to 36.35 s for Q38FN and 83.87 s for GLM53F. If you mean two independent instances of DeepSeek, one per
You have reached the end of the archive
All of deepseek