Field notes ยท from two hands-on evals

Saving money with Fable 5

For agents Reading this with an AI agent? Point it at the markdown version instead: fable-5-news.pages.dev/savings.md

Fable 5 is the most capable Claude model - and the priciest way to burn a Claude subscription. Two hands-on evals (a solo-vs-delegation orchestration test, and a coding-subagent shootout) point to a simple playbook for using it economically. Takeaways first, evidence below.

Same feature. Same workers. Different manager.

Claude spend to manage one real coding task through the same pool of cheap workers

Fable 5 managesno optimization $6.42
Opus managesthe one-line change $2.13
−67% Claude spend, with every quality check still passing. The catch: it's one measured run, and the job takes longer than not delegating at all - details below.
~⅓

of solo Claude-plan burn when Fable delegates the grind to a flat pool of cheaper models

$2.13

management cost with Opus as the flat manager - cheaper than Opus doing the work solo ($4.69–$5.23)

~40%

faster: solo beat flat delegation on wall clock at this task size (614s vs 1044s avg)

All numbers from the two eval reports summarized below; small sample sizes throughout (n=1–3 per cell).

The three setups, in one picture

"Solo," "flat," and "nested" are three ways of getting the same coding task done. The same real feature was built under each shape and measured on time, cost, and quality.

Opus 4.8 does all the work itself Solo

One strong Claude model works alone: reads the spec, writes the code, runs the tests, commits.

manager gpt gpt glm glm one hop: manager → workers Flat delegation

A Claude manager (Fable 5 or Opus) splits the task and hands pieces directly to cheaper workers on other subscriptions, then integrates and verifies.

Fable 5 Opus mgr Opus mgr glm glm gpt glm Nested delegation

Fable 5 delegates to Opus middle managers, who delegate to workers. Two hops between the spec and the code.

The workers throughout: GPT 5.5 (via codex, on a ChatGPT plan) and GLM 5.2 (via opencode, on a Z.AI plan) - so their token grind lands on those subscriptions, not the Claude plan.

Five ways to spend less

So where does Fable 5 fit?

Honest answer: in these two evals, Fable 5 never won a measured category - but it was only ever tested as a manager, the one place its extra capability can't pay for itself. Routing work doesn't need the smartest model in the room; it needs a cheap, reliable one. Fable's case is the task itself being too hard for everything else. At a glance:

The task is genuinely harddeep reasoning, huge context, everything else has failed Fable 5, working alonepay for capability where capability is the bottleneck (untested in these evals - they never ran Fable solo)
Normal task, you want it fastsmall feature, bug fix, refactor Opus 4.8, working alonefastest measured shape at this task size
Normal task, protect your Claude planlots of tasks, limited weekly usage Opus manages cheap workerscheapest Claude spend measured ($2.13); grind lands on other plans
Any task at all Fable as manager, or managers-of-managerswon nothing in either eval

The evidence

How this was tested

One real feature (a preset pack/unpack command plus UI) was built under each orchestration shape, from frozen prompts in isolated git worktrees, with identical acceptance checks: the test suite, a typecheck, a byte-identical CLI round trip, and a UI render check. Every cell passed all four. A blind GPT 5.5 judge then scored each diff against the spec. Solo ran twice, Fable-managed flat twice, Opus-managed flat once, nested once.

Wall clock

Seconds to complete the feature, lower is better

Solo Opus, avg of 2614s
Flat Opus manager, n=1904s
Flat Fable manager, avg of 21044s
Nested Fable → Opus x2, n=11353s

Both delegated shapes ran workers concurrently and still lost to one model alone - overhead beat parallelism at this task size.

Claude cost, management layer

Dollars of Claude usage; workers ride other plans, so this is not a total-system claim

Flat Opus manager$2.13
Solo Opus$4.96
Flat Fable manager$6.42
Nested$7.87

Managing the worker pool costs less than doing the work at the same model's prices - if the manager is Opus, not Fable.

Claude-plan burn

Output tokens across all Claude layers per run, lower preserves more subscription headroom

Flat Opus manager~22k
Flat Fable manager~30k
Solo Opus~39k
Nested~76k

Cache reads tell the same story: solo ~6.1M vs flat ~1.2–2.3M. Nested forfeits the headroom win - its Opus managers put the grind back on the Claude plan.

Show the full data table
CellWall clockClaude cost (mgmt)WorkersFidelity (blind, 1–5)
Solo (Opus, avg of 2)614s$4.96none5
Flat (Fable manager, avg of 2)1044s$6.422 gpt + 2 glm4
Flat (Opus manager, n=1)904s$2.132 gpt + 2 glm4
Nested (Fable → Opus x2, n=1)1353s$7.871 gpt + 6 glm4

Plan burn per cell (all Claude layers): solo ~80 turns / ~6.1M cache reads / ~39k output; Fable-flat ~44 / ~2.3M / ~30k; Opus-flat 31 / 1.2M / 22k; nested 26 / 4.6M / 76k.

The "solo is higher fidelity" gap did not survive a code audit.

The blind judge scored solo 5/5 and every delegated run 4, but reading the diffs directly found the same two deviations (non-atomic unpack, no export-set selector) present in the solo runs too - just unflagged. Every run, solo or delegated, skipped real atomicity; nothing was "lost in the delegation telephone game" except one verified nested-only defect. Treat the fidelity gap as judge noise pending a re-judge. Also: nested and the Opus-manager cell are n=1, and Opus ran at default effort vs Fable at high, so effort is conflated with model choice.