Mistral Medium 3.5 Code Test Feedback
Jorlen · reddit · 2026-07-10
The poster asks if anyone has used Mistral Medium 3.5 (128B dense) in code projects, and shares their own trial of Unsloth's 3-bit quantized version Q3KS.
They say they ran it for a few hours on their latest project, and the model found issues missed by daily drivers; though slower than MoE, it's acceptable for their setup, about 8 tok/s, 80k context.
More from Models
- Claim says Kimi was distilled from Fable, sparking a model-attribution jab — cephaloform · 2026-07-22
- Gary Marcus says LLM math skills are like knowing only a car’s engine size — GaryMarcus · 2026-07-22
- OpenAI’s Codex + GPT-5.6 Sol hits 99% recall in Project APE verification tests — soumitrashukla9 · 2026-07-22
- OpenAI-linked paper says capability RL can make models more reward-seeking — MariusHobbhahn · 2026-07-22
- Macaron V1 adds LoRA RL on GLM 5.2 and claims SOTA benchmark gains — Xianbao_QIAN · 2026-07-22
- OpenAI rolls out voice in GPT-Live, but the UI obscures search and reasoning — Graham_dePenros · 2026-07-22