Models

Pinned Qwen 3.6 35a3b & 27b are incredible for small GPUs, Gemma 4 26a4b & 31b are runners-up
by Nick Antonaccio

Pinned Deepseek V4 Flash is currently my favorite local model, without reservation (and great on the API too)
by Nick Antonaccio

Pinned Inexpensive LLM API providers for open source models
by Nick Antonaccio

Pinned Kimi K3
by Nick Antonaccio

Pinned Hy3 works locally on a single DGX Spark. It's a serious contender for best self-hosted model.
by Nick Antonaccio

Ternary models
by Nick Antonaccio

arcee-ai/trinity-large-thinking
by Nick Antonaccio

Don't forget about Deepseek
by Nick Antonaccio

google/gemini-3.1-flash-lite-preview
by Nick Antonaccio

Qwen 3.6 Plus vs Recent Leaders
by Nick Antonaccio

Use MTP!
by Nick Antonaccio

Minimax M3 - my first test results with it are *Fantastic*
by Nick Antonaccio

Laguna S 2.1 is the newest great mid-size self-hostable model.
by Nick Antonaccio

Don't forget about GLM 5.2
by Nick Antonaccio

Why use self-hosted models?
by Nick Antonaccio

Testing LLMs' depth of knowledge, all the way up to Anthropic Fable
by Nick Antonaccio

Step-3.7-Flash is surprisingly good
by Nick Antonaccio

Tencent/hy3-preview is the current medium sized model to watch
by Nick Antonaccio