Pull down to go back
Huge performance boost: switching from Ollama to llama.cpp actually makes a difference
opensourcereddit

Huge performance boost: switching from Ollama to llama.cpp actually makes a difference

從 Ollama 換到 llama.cpp 後效能大幅提升，機器人對戰結果真的不一樣

Someone built tiny AI-controlled robots that fight each other to survive, and here's the wild part: when they switched from Ollama to llama.cpp and had Qwen3 Coder generate the robot battle code, the robots started winning more matches consistently. It's basically watching an AI learn to fight better in real-time. The creator made it open-source on GitHub, and apparently it's genuinely fun to tweak the parameters and see how different settings change the robots' performance. If you're curious about LLMs actually doing something practical (and entertaining), this is worth checking out.

Read original →

Keywords

llama.cppollamaperformance improvementcode generationrobot simulationopen source

Related Articles

TRELLIS.2 image-to-3D now runs on Mac (Apple Silicon) - no NVIDIA GPU needed
Open SourceRead

TRELLIS.2 image-to-3D now runs on Mac (Apple Silicon) - no NVIDIA GPU needed

I ported Microsoft's TRELLIS.2 to run on Apple Silicon via PyTorch MPS. The original depends on five CUDA-only compiled extensions (flex_gemm, flash_attn, o_voxel, cumesh, nvdiffrast) that have no Mac equivalent. Wrote replacement backends from scratch: Pure-PyTorch sparse 3D convolution (replacing flex_gemm), Python mesh extraction using spatial hashing (replacing CUDA hashmap ops in o_voxel), SDPA attention for sparse transformers (replacing flash_attn), and GPU-accelerated trilinear interpolation (replacing cumesh and nvdiffrast).

Reddit
Show HN: Run TRELLIS.2 Image-to-3D generation natively on Apple Silicon
Open SourceRead

Show HN: Run TRELLIS.2 Image-to-3D generation natively on Apple Silicon

I ported Microsoft's TRELLIS.2 (4B parameter image-to-3D model) to run on Apple Silicon via PyTorch MPS. The original requires CUDA with flash_attn, nvdiffrast, and custom sparse convolution kernels: none of which work on Mac. I replaced the CUDA-specific ops with pure-PyTorch alternatives: a gather-scatter sparse 3D convolution, SDPA attention for sparse transformers, and a Python-based mesh extraction replacing CUDA hashmap operations. Total changes are a few hundred lines across 9 files.

Hacker News
157
Which Qwen 3.6 Quantization Works Best for M5 Pro with 24GB RAM?
Open SourceRead

Which Qwen 3.6 Quantization Works Best for M5 Pro with 24GB RAM?

A user with an M5 Pro machine featuring 24GB of RAM is trying to figure out the best quantization level for running Qwen 3.6 via Ollama. They're unsure whether to go with Q4 (4-bit quantization) but can't find a solid Q3 (3-bit) alternative. They're looking for recommendations on which version would run smoothly on their setup without compromising too much on quality.

Reddit