This 106M-param model plays Doom at 10ms on a single CPU core, beating Jev
Playful_Suggestion_3 · reddit · 2026-09-29
The author released indecis, an MIT-licensed Go library that fine-tunes a small text encoder (106M params, tens of MB) to answer typed questions with probabilities, letting it play Doom at 10ms median latency on a single CPU core — versus Jev's 70–500ms hosted decisions and LAYA's 421M params at 98ms. Across 20 ViZDoom defendthecenter rounds it scored +21.8 points on average, matching the scripted bot that generated its training data, using 65MB of RAM and just 11 minutes of laptop training. The real goal: replacing the LLM-based request classifier in the author's Xolo gateway to cut hundreds of ms of network round-trip.
More from Models
- Opus 5.5 vs Sol 6.1: Drawing Mona Lisa as Pure SVG From Memory, Both Deeply Derpy — OriginalScrubLord · 2026-09-30
- MiMO ships an agentic critic; early speedups but long-run bias concerns — teortaxesTex · 2026-09-30
- Intern-Decision open models (0.8B–4B) beat Jev on multimodal decision benchmarks — max_paperclips · 2026-09-30
- No reason for OpenAI to keep both $100 and $200 plans, argues Redditor — Over_Sheepherder4503 · 2026-09-30
- Heavy user proposes pairing OAI Pro with a 512GB Mac Studio running GLM-5.3 locally — paperchow · 2026-09-30
- Altman 'Will Think About' More Open Models; Community Points to NVIDIA Nemotron Instead — omarsar0 · 2026-09-30