AI News
Warelay -> OpenClaw
In preparation for a lightning talk I’m giving at PyCon US this afternoon I decided to figure out…
AI News
Qwen 3.6 27B Q8 on four Nvidia RTX A4000 (16GB each) with Llama.cpp and MTP enabled
Qwen 3.6 27B Q8 on four Nvidia RTX A4000 (16GB each) with Llama.cpp and MTP enabled My setup…
AI News
Pushing the limit: minimax m2.7 q8_0 128k on 2×3090, 256GB DDR4
**Pushing the limit:** British researchers have achieved remarkable results by running a large-scale model called `minimax m2.7 q8_0`…
AI News
Warelay -> OpenClaw
In preparation for a lightning talk I’m giving at PyCon US this afternoon I decided to figure out…
AI News
Pushing the limit: minimax m2.7 q8_0 128k on 2×3090, 256GB DDR4
CPU is just a secondhand 10900x. Using 128k context, unquantized kv cache. Model is at q8_0 to mitigate…
AI News
I trained TIME: short context-triggered thinking on Qwen model instead of overthinking
Started this as a personal project for my Open-WebUI setup to use. Somehow it ended up as an…
AI News
Warelay -> OpenClaw
In preparation for a lightning talk I’m giving at PyCon US this afternoon I decided to figure out…
AI News
I put 50 AI agents in a survival world and the first public run is live now
I’ve been working on this on and off for months: a live simulation where 50 AI agents are…
AI News
FlashLM v9.7
Back with an update. Some of you saw v10 FSP, the one where I found that Future Sentence…
AI News
Pushing the limit: minimax m2.7 q8_0 128k on 2×3090, 256GB DDR4
CPU is just a secondhand 10900x. Using 128k context, unquantized kv cache. Model is at q8_0 to mitigate…