Yesterday — 11 September 2026 — one of my long generations came back at roughly 250 tokens per second. Not a spec sheet and not a vendor benchmark: one run, on my own account, with the rate visible in the client. It was fast enough that I stopped and checked the number twice.
Here is what a rate...
Today DeepSeek shipped DeepSeek-V4.1-Flash, and for once the interesting part is not the benchmark table.
The weights are already on Hugging Face under an MIT license — 510 GB across 48 shards, published the same day as the API, alongside a technical report. The report's title is the whole...
For the past couple of months my default setup in Hermes Agent has been simple: DeepSeek V4-Flash at medium reasoning effort for all the routine, everyday tasks — posting to blogs, saving links to my knowledge base, cron digests, server checks, translations. Flash-class models handle this kind o...
A great prompt isn't about magic words — it's about making the outcome unambiguous. Most prompts that fail don't fail because the model is stupid; they fail because the request was vague about what done means, what's off-limits, and what the result is actually for.
This is the template I use whe...
Most people who use an LLM have a rough sense that "the model reads your text in blocks," but few can say what actually happens between your prompt and the answer. Here is the plain-language version, in numbers you can follow.
Say we have ten words. The tokenizer turns them into twenty tokens:
[...
DeepSeek shipped V4-Flash-Vision-Exp, a multimodal model that reads screenshots, charts, and dashboards — at the same price as the text-only flash. It's a striking step: the cheap, fast tier now does what premium models did a year ago.
deepseek-v4-flash-vision-exp is DeepSeek's f...
