My end goal is to give an LLM a Docker container and the ability to run shell commands and be able to have it Do Stuff there – edit files, run commands etc – but I know that's wildly ambitious
@simonw
-
Running Devstral on 64GB M2 Mac with Ollama
By
–
64GB M2 Mac, might be able to run Devstral well on it, I only tried that in Ollama so far, need to try other harnesses to see if they support tools more reliably
-
Qwen Code Model Exceeds Mac Memory Capacity Limitations
By
–
Sadly the Qwen Code model doesn't fit on my Mac! I only have 64GB
-
Models Making Multiple Tool Calls Single Conversation
By
–
It's really the number of tool calls you can make a single conversation Many models can do one just fine, but I want to be able to do a dozen or more
-
GPT 4.1, o3, o4-mini and Claude 4 comparison feedback
By
–
I have been happy with GPT 4.1 and o3 and o4-mini for this in addition to Claude 4
-
Local Model Under 30GB Memory for Tool-Calling Tasks
By
–
I'm trying to find a local model that uses less than about 30GB of memory to run (since my M2 has 64GB and I want to run other stuff too) that's competent at running tools in a loop to power something similar to Claude Code
-
Local Model Agentic Coding Loop Tool Calling Challenge
By
–
Is anyone having success running a full Claude Code style agentic coding loop against a local model? This feels like the ultimate challenge for local tool calling right now, as it requires potentially dozens of calls in a loop and good performance over a longer context
-
Datarama tool reminds of similar AI data solutions
By
–
Reminded me a bit more of this one! https://
simonwillison.net/2025/Jun/11/da
tarama/
… -
Timing Build Derivation from Commit History Analysis
By
–
You can derive the timing of the build from the commit history https://
github.com/simonw/system-
exploration-g/commits/main/
…