Local LLM for agentic work on 16GB VRAM: gpt-oss-20b refuses too much, what else?
SLTGAMER · reddit · 2026-10-21
A Reddit user asks for local model recommendations for agentic workflows on 16GB VRAM + 32GB RAM. They tried many models: gpt-oss-20b performs well but has a high refusal rate; a Qwen 3 8B variant felt slow and a bit weak. They've been using Hermes and are open to suggestions.
More from Models
- Claude Max now includes $100/month in API credits, screenshot hints at Opus 5.5 with 1M context — michalmalewicz · 2026-10-21
- OpenRouter listing shows GPT-6 Sol outpacing larger GPT-6 Luna, sparking debate — tinteh · 2026-10-21
- Full video: Pichai on Argon's 25 trillion tokens in 72 hours now on YouTube — rohanpaul_ai · 2026-10-21
- Is that "open-source" AI actually open? A 4-level framework, three models scored — alex_verem · 2026-10-21
- InternScience releases Sage-A1.5, a 35B MoE agentic model fine-tuned from Qwen3.6-35B-A3B — Skyline34rGt · 2026-10-21
- 21 LLMs, 7,980 code reviews: higher reasoning effort often reduced precision — Electrical_Walrus537 · 2026-10-21