I recently switched to using for some large volume inference with small, fine-tuned models and was surprised how nice and painless things were. vLLM seems to have a lot more gotchas and kludged together stuff once you get outside of anything straightforward.
I’m old enough to remember the heavenly times before people repeated phrases like “the taxi cartel” as if they were original thoughts vs. cooked up by Uber’s comms team.
I forgot this was in effect. I’ve been hitting my weekly rate limits two to three days in because of needing to steer Slopus 5 with Fable. Absurd that they’re being cut further in the face of steep competition from open weights models.
It’s not “deeply weird” but was intentionally legislated to prevent labor from obtaining similar sectorial bargaining power and political influence as they had in Europe. Some entertainment industry unions had already achieved similar stature pre-WWII and were grandfathered in, but the US learned their lesson here and explicitly sought to atomize at the workplace level.
Use their API directly, this is an openrouter issue. I ran something like 5 billion tokens through them directly recently without any bumps in the road.
Deepseek has demonstrated that there is no reason for it to actually lose money. The awful business practices and monopoly tactics of the frontier model labs in the US are the problem.
Tell me you know nothing about Switzerland without telling me you know nothing about Switzerland. Try asking a German Swiss what they think about a French Swiss or either about the Romansch.
Bacurau is one of the best movies I’ve seen in recent memory and Pictures of Ghosts tells an amazing story about the history of Recife’s relationship to cinema.
Unfortunately, this looks to only cover the larger MoE models. I imagine the smaller models are what most people would target. 9B just dropped two days ago, so not surprised it’s not explicitly documented, but does use a hybrid mamba architecture that I expect needs some special consideration.
Ireland’s affordability problems are almost exclusively centered around its housing crisis and they need to just commit themselves to over-supply induced wealth destruction for the landlord class and older generations. Thankfully, there demographics also support such a move.
I tested this pretty extensively and it has a common failure mode that prevents me from using: extracting footnotes and similar from the full text of academic works. For some reason, many of these models are trained in a way that results in these being excluded, despite these document sections often containing import details and context. Both versions of DeepseekOCR have the same problem. Of the others I’ve tested, dot-ocr in layout mode works best (but is slow) and then datalab’s chandra model (which is larger and has bad license constraints).
Yeah, do crons even work consistently for GitHub Actions? I tried to set one up the other day and it just randomly skipped runs. There were some docs that suggested they’re entirely unreliable as well.
Everyone who tells the story of the reformation leaves out that Martin Luther also used this new technology to widely disseminate his deranged anti-Semitic lies and conspiracies, leading to pogroms against Jews, a hundred years of war across Europe, and providing the ideological basis for the rise of Nazism.
Tried this the other day and the setup on this is super cumbersome and requires you to constantly rebuild your entire dev and Claude Code environment every time you use a new container, including whitelisting URLs for package managers and the like.