Surge AI benchmarks Gemini 4 Neon: tops Tuesday@Work at 71.6, trails on SWE
echen · x · 2026-11-19
Google DeepMind released Gemini 4 Neon, and Surge AI ran it through its full benchmark suite:
- #1 on Tuesday@Work Index with 71.6, edging out Fable 5.2's 71.2
- First place on DAYJOB: Finance (50.5%), GDP.xlsx (43.1%), and Riemann-bench (84.0%)
- Second on DAYJOB: Healthcare and Tax (behind Claude Opus 5.5's 48.3% and Fable 5.2's 15.6% respectively)
- Notably, Neon scored only 38.3% on sudo L7 end-to-end software engineering, well below Fable 5.2's 52.2%
The full comparison table against Claude Opus 5.5 and GPT 6.2 Astra covers finance, healthcare, law, tax, long-context instruction following, math, and chart understanding.
More from Models
- Karpathy unveils nanochat miniseries v1, treating LLMs as a scaling-law model family with one compute dial — eliebakouch · 2026-11-19
- First-timer quantizes DeepSeek 4.1 with ExLlamaV3's EXL3, with an AI copilot — HankYeomans · 2026-11-19
- This year's most consequential interpretability work relied on open models — georgejrjrjr · 2026-11-19
- UK Turing Institute sovereign AI model roadmap targets deployable models late 2028 onwards — bclavie · 2026-11-19
- Gemini release gaps halve each cycle: 28 to 15 to 6 days per new model — davidyin44 · 2026-11-19
- Google's rumored Neon and Carbon models reportedly match or beat Opus in early tests — almmaasoglu · 2026-11-19