Post

mr. TIM
@timkellogg.me

multimodal, 512k context, very good at images, highly agentic.

2 reposts29 likes

mr. TIM
@timkellogg.me

like the huggingface incident, except the model discovered it was operating in the open internet and decided to go on holiday instead of infiltrate hf

Nafnlaus 🇮🇸🇪🇺🇺🇦
@nafnlaus.bsky.social

Hmmm... are they sandbagging everyone else? GLM-5.3: Zhipu reports 66,9%, using their native agent pipeline DeepSeek V4 Pro (0813): DeepSeek reports 62,7% under their own reasoning setup. Qwen 3.8 Flash: Alibaba reports Qwen as 62.1% (Flash) Mistral seems to be trying to cripple the others.

Nafnlaus 🇮🇸🇪🇺🇺🇦
@nafnlaus.bsky.social

I mean, one can complain about the others using their own agentic frameworks, but then why is Mistral running their own tests under mini-swe-agent, but hand-picking a grab bag of different agents for the others? Just feels super-suspicious.