Fable was refusing to patch vllm for me when trying to get mtp to work on r9700 gpus. Kept on bumping down to opus. Tried to really sanitize my prompts and everything but it seemed intrinsically prohibited from doing this sort of work. I guess it’s useful for making inane one shot games and websites, lol.

▲

mrandish 8 hours ago | parent | next [-]

> Tried to really sanitize my prompts

And in doing so, you probably got your account and prompt flagged for 'attempted jailbreaking' (apparently, such scores are remembered for up to 7 years).

▲

bconsta 5 hours ago | parent [-]

Kind of ironic considering Anthropic hasn't even been around for 7 years.

	▲	mrandish 5 hours ago \| parent [-]
		Yeah, I added that because someone else posted the wording of an Anthropic disclosure saying they'll keep your data for two years and security data for 7 years.

▲

wolttam 10 hours ago | parent | prev | next [-]

I was recently using self-hosted DeepSeek V4 Flash to poke around the DSpark implementation in vLLM (well outside of my domain)

I did wonder if I was doing anything Fable would have flagged - sounds like yes.

	▲	fer 10 hours ago \| parent [-]
		Fable refuses to touch anything in my side project because it uses libtorch. It will bounce even for parts that have nothing to do with it.

▲

pmdr 10 hours ago | parent | prev | next [-]

That's where the free marketing comes from.

▲

_davide_ 10 hours ago | parent | prev | next [-]

same experience here, as soon as it touched any gpu code it stopped working

	▲	thx67 10 hours ago \| parent [-]
		I had it refuse OpenCL code that had nothing to do with anything, just slightly advanced cellular automata.

▲

chews 10 hours ago | parent | prev [-]

Anthropic's TOS clearly says they don't want to facilitate any sort of distillation, it's not a stretch to think they will limit any sort of learning on improving other models.

▲

echelon 10 hours ago | parent [-]

Literally pulling the ladder up.

Disgusting behavior.

I like the product, I hate the company. I can't wait for competition.

▲

christophilus 7 hours ago | parent [-]

Competition is here. I personally prefer Codex. Opencode with a variety of Chinese models is also just fine for 80% of my use cases.

	▲	s1gsegv 2 hours ago \| parent [-]
		What are people doing to maximize the value out of workflows like this? I’ve struggled to integrate other models into my workflow because you get so much Opus for the $100 Max 5x plan, then there’s no step down with Claude Code access. Some other plan people like?