Enterprise is not freaking out because DeepSeek 4.1 Flash does not actually occupy a spot on the Pareto frontier for non-coding enterprise workflows. We see this at my employer, focused on non-technical knowledge work. Luna 6 and now Haiku 5.5 are both very competitive if not better on all axes that we care about

Because most of the people use it through enterprise agreements and don't pay the bill? I run it for my own use cases and its pricing plus caching capabilities are hard to beat, cents for millions of tokens. https://substack.com/@rubenafo/note/c-332218129?r=26y5kn&utm...

Why ?

Because I don't have the time and money to do an extensive benchmark of all major LLM, so when I had to select a LLM for my usage (which was not coding at first), I went to the most used one, chatgpt, because I knew if would be one of the best at the task.

I suspect it's the case for many if not most people.

Been using Flash 4.1 via the ante harness to blast through a GBA recomp. The ante team has pushed hard to make Flash 4.1 perform well under it. So far, I've maybe spent $10 over the last 3 days. Its a real workhorse and works much better in this harness

It's like the boy crying wolf with these models, Deepseek is not as good as Claude, I like using Gemini Flash Lite even, it is good enough for most of my crud app tasks, brownfield projects you don't need the strongest models, but greenfield projects with lack of context from indexed code and lack of domain expertise, I think the Opus models have been working for people

> This is like comparing big pharma with generic manufacturers who can skip the R&D. While these drugs are not 1 to 1 copies, we are still comparing apples to apples, but with a 90% price cut.

Weird! It's almost like distillation is bad for the long-term growth of the industry, just like generic manufacturers would be if they could release the generic versions of drugs 2 weeks after the original R&D completes.

Very strange sentence to include in an article after saying "I don't care at all about distillation" up at the top. Can the author not hear themselves?

> China is going to eat their lunch

No doubt about it, that's why their push for international regulation to the levels of nuclear inspections using the narrative of annihilation and apocalypse

Because it’s not even that cheap? The author chose to only include Claude in their chart and ignored the fact that 6.1-sol and even more so luna can easily beat Deepseek on cost. Of course almost free cache used to be the main differentiator, raw token cost is deceptive since 4.1 just uses way more tokens than most other models

DeepSeek 4.1 Flash 0910 is perfect for M5 Ultra 256GiB. Running it fully resident in RAM, prefill at ~2500 tok/s and decode at ~40 tok/s. Probably tons of room to improve from there.

> Sure, they stole Claude's training, and Anthropic stole it from other people. I'm not getting into the whole who-owns-whose-data debate, because most developers aren't thinking like that. They're just trying to get the most bang for their buck.

Until they get laid off and suddenly discover their moral compass.