Surge AI Launches Tuesday Work Index; Fable 5 Edges Out GPT 5.6 Sol
On August 19, Surge AI launched Tuesday (The Frontier Work Index), a new benchmark scoring AI models on professional work, and published its first single-score leaderboard: Fable 5 tops it at 66.8, with GPT 5.6 Sol close behind at 66.5. The benchmark targets real workplace competency, marking a shift from static academic questions toward simulated real workflows.
Confirmed
- Tuesday assumes "professional intelligence is not one skill" and combines 8 sub-benchmarks into a single score: Chartography, HANDBOOK.md, GDP.pdf, ComplexConstraints, CoreCraft, Hemingway-bench, Antidote, and Rie (name truncated in posts), with more to be added.
- It simulates a typical Tuesday morning: reading emails, understanding the data team's charts, connecting last week's Slack threads, and reporting to a VP before noon.
- Skills tested include reading charts, following policy, tracking constraints across channels, tool use, and explaining business insights to a boss.
- First leaderboard: Fable 5 at 66.8, GPT 5.6 Sol at 66.5, then a clear gap — DeepSeek V4 Pro at 59.7, Qwen 3.8 Max at 58.7, followed by Gemini and others (later scores truncated).
- @echen added that models need both breadth (long-horizon work) and narrow skills (chart reading), plus soft skills: a perfectly reasoned but impenetrable report nobody wants to read; inspiring output requires judgment and taste.
Why it matters
- The top two differ by only 0.3 points, showing leading models are nearly tied on workplace competence, while a near 7-point gap to the second tier indicates clear stratification.
- By synthesizing breadth, narrow skills, judgment, and taste into one comparable score, Tuesday offers the industry a new horizontal evaluation lens.
2026-08-19 ~ 2026-08-20 · 7 related posts
Primary sources
- Surge AI Launches Tuesday Index to Benchmark AI on Real-World Professional Tasks — echen · 2026-08-19
- Tuesday Work Index leaderboard: fable 5 tops at 66.8, gpt 5.6 sol at 66.5 — echen · 2026-08-19
- [source] Tuesday Leaderboard: Fable 5 and GPT 5.6 Sol Lead the Pack — echen · 2026-08-19
- Surge: professional intelligence needs breadth, narrow skills, and taste — echen · 2026-08-19
- [source] Tuesday index combines 8 Surge benchmarks, more coming — echen · 2026-08-19
- [source] Surge AI launches Tuesday Frontier Work Index, 8 benchmarks in one score — echen · 2026-08-19
1 near-duplicate retellings: echen