Testing Local Models for Excel Tasks
TechCrunch AI · rss · 2026-07-14
The author tested local models for Excel-related tasks, finding Deepseek v4 Flash (used via DS4) to offer the best experience so far.
They noted that the model runs at roughly 30–40 tk/s on moderate context lengths with solid overall output. The author is also looking for other solutions suitable for running on a Mac Studio with 256GB RAM.
More from Models
- Sol61 flunks instruction-following: skips mandated research even when explicitly told — lucasmeijer · 2026-10-26
- Kimi K3.1 Pro tops MathArena open models, closing frontier gap to about 6 months — teortaxesTex · 2026-10-26
- Ruff creator Charlie Marsh: frontier models find infinite bugs in any software — charliermarsh · 2026-10-26
- Dev testing one skill across Opus 5.5 and Sol 6.1 finds instruction following 'all over the map' — lucasmeijer · 2026-10-26
- supermemory unveils learner-1, claiming continual learning for any model — julianweisser · 2026-10-26
- Anthropic says Opus 5.5 cuts cost per task by ~40% vs Opus 5 — taufiqintech · 2026-10-26