AI Briefing#local-llm
The latest AI and developer news about #local-llm, with the original source and a short summary.
The latest AI and developer news about #local-llm, with the original source and a short summary.
On an RTX6000 Blackwell, Qwen 3.6 27B BF16 generated a Pac-Man HTML.
The little-coder and Qwen3.6-35B-A3B combo scored 78.7% on Aider Polyglot.
In a 108K-token JS file memory test, Qwen 3.6 significantly outperformed Gemma 4.
Roo Code is shutting down after surpassing 3 million installs, shifting full focus to Roomote.
Even with 32GB VRAM, llama.cpp's `--fit` ran Qwen3.6 Q8 at a 256k context at 57 t/s.
Running the same prompt across 9 local models revealed a correlation between quantization provider differences and code length.
Trained a 235M-parameter LLM from scratch using a single RTX 5080.
This is a benchmark result running MiniMax M2.7 4bit on dual Asus GX10 at about 100W.
Shared a setup for running Qwen3.6-35B-A3B with llama-server on an RTX 4060 8GB VRAM, along with a thinking budget pitfall.
In Terminal-Bench 2.0, MiniMax M2.7 scored 41.3%, lower than M2.5.
An experiment tackling SWE-bench Verified by running Qwen3.6-35B A3B on 2x Tesla P40.
v1.4.0 shrinks the APK and adds in-app model download and switching.
With the same Qwen3.5-9B Q4, simply changing the scaffold pushed Aider performance from 19.11% to 45.56%.
Llama.cpp token speed measurements on the Zenbook A16 with Snapdragon X2 Elite Extreme.
A 35-domain LoRA routing model based on Qwen3.5-35B-A3B has been released.
On the RTX 5060 Ti 16GB, Qwen3.6-35B-A3B ran about 10% faster than the prebuilt version.
Combining an RX 9070 and RTX 5070 Ti, the author ran Qwen 35B Q4 at over 100tps.
In a 30k LOC agentic evaluation, Qwen 3.6 35B outperformed Gemma 4 26B.
On dual RTX 5060 Ti GPUs, Qwen 3.6-35B-A3B hit 21.7 tok/s at 90K context.
In the Qwen3.6-35B-A3B GGUF benchmark, Unsloth quants claim the best size efficiency relative to KLD.
Qwen3.6-35B-A3B-UD-Q2_K_XL handled 58 tool calls at a 98.3% success rate on a 16GB VRAM laptop.
A locally run Qwen3.6-35B-A3B drew a better pelican SVG than Opus 4.7.
An abliteration version targeting the MoE characteristics of Qwen3.6-35B-A3B has been uploaded to HF.