After testing Opus 5 and Fable 5 across 15 projects, one user says Opus wins most work
doodlestein · x · 2026-07-25
The author says that after using Opus 5 and Fable 5 across roughly 15 projects, they would switch to Opus for most work.
Their reasons are pragmatic: Opus appears to be faster, more persistent, and more agentic without giving up much intelligence. They still prefer Fable for tasks that depend on broad world knowledge and extreme long-tail facts—such as detailed history, biological trivia, novel summaries, or the internals of open-source projects over time.
In practice, they plan to use Fable mostly for drafting, reviewing, and revising markdown plans, while using Opus for implementation work when the prompt and skills are sufficiently detailed. They also argue that the best results come from combining models rather than using any single one in isolation, citing a workflow where Sol Max, Opus 5, agent mail, and ntm review each other’s work.
Related event: Opus 5 vs Fable 5: Users Debate the Better Model(3 posts)→
More from Models
- GPT-6 Astra beats Factorio with enemies in 44 in-game hours at ~$4,500 API cost — liminal_bardo · 2026-09-11
- giffmana: the env being used in training is part of the point — giffmana · 2026-09-11
- awesome-llm-leaderboards: an open-source directory of LLM leaderboards, pricing tables, comparison tools — Last_Establishment_1 · 2026-09-11
- Anthropic claims it works to keep eval environments unidentifiable to models — MaxKannen · 2026-09-11
- Nex N2.5 Pro released on Hugging Face with 407GB of weights — jinnyjuice · 2026-09-11
- RoMa v2 image matching model unveiled in the usual black poster — ducha_aiki · 2026-09-11