mirror of
https://github.com/fabro-sh/fabro.git
synced 2026-09-17 23:52:34 +00:00
a0fbbcfe3b
This PR adds support for Anthropic's fast mode beta feature (`speed:
fast`), which enables up to 2.5x faster output for Claude Opus 4.6 at a
6x pricing premium. The implementation follows the same patterns
established by `reasoning_effort`, threading a new `speed:
Option<String>` field through the request/response pipeline from graph
stylesheet properties down through agent configuration, session
management, and the Anthropic provider adapter.
On the provider side, when `speed: "fast"` is set, the `ApiRequest`
struct now includes the `speed` field in the serialized JSON body, and
the `build_beta_header` function injects the required `anthropic-beta:
fast-mode-2026-02-01` header alongside any existing beta headers (cache,
interleaved thinking, etc.) without duplication. The response's
`usage.speed` field is parsed and propagated back through both streaming
and non-streaming paths into `StageUsage` and `Usage` types for
tracking.
Cost accounting applies a 6x multiplier in `compute_stage_cost` when
`speed == "fast"`, reflecting Anthropic's actual pricing differential.
The feature is configurable via stylesheet (`* { speed: fast; }`), which
gets wired through `SessionConfig` and prompt-mode `Request`
construction in the backend layer. New tests cover the API request
serialization, beta header injection, combined cache+fast-mode headers,
and the cost multiplier, while all existing test fixtures have been
updated with `speed: None` to maintain struct exhaustiveness.
### Fabro Details
<details>
<summary>Ran 9 stages in 58m 27s for $6.56</summary>
| Stage | Duration | Cost | Retries |
|---|---|---|---|
| start | 0s | – | 0 |
| toolchain | 0s | – | 0 |
| preflight_compile | 1m 20s | – | 0 |
| preflight_lint | 14s | – | 0 |
| implement | 47m 31s | $5.36 | 0 |
| simplify_opus | 7m 59s | $1.20 | 0 |
| simplify_gpt | 0s | – | 0 |
| verify | 25s | – | 0 |
| fmt | 1s | – | 0 |
| **Total** | **58m 27s** | **$6.56** | **0** |
</details>
<details>
<summary>Ran <code>ImplementAndSimplify.fabro</code> (12 nodes and 15
edges)</summary>
```dot
digraph ImplementAndSimplify {
graph [
goal="Implement and simplify",
model_stylesheet="
* { backend: api; model: claude-opus-4-6;}
"
]
rankdir=LR
start [shape=Mdiamond, label="Start"]
exit [shape=Msquare, label="Exit"]
toolchain [label="Toolchain", shape=parallelogram, script="command -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", max_retries=0]
preflight_compile [label="Preflight Compile", shape=parallelogram, script="cargo check -q --workspace 2>&1", max_retries=0]
preflight_lint [label="Preflight Lint", shape=parallelogram, script="cargo clippy -q --workspace -- -D warnings 2>&1", max_retries=0]
fix_lints [label="Fix Lints", prompt="The preflight lint step failed. Read the build output from context and fix all clippy lint warnings.", max_visits=3]
implement [label="Implement", prompt="Read the plan file referenced in the goal and implement every step. Make all the code changes described in the plan. Use red/green TDD."]
simplify_opus [label="Simplify (Opus)", prompt="@prompts/simplify.md"]
simplify_gpt [label="Simplify (GPT-54)", prompt="@prompts/simplify.md", model="gpt-54"]
verify [label="Verify", shape=parallelogram, script="cargo clippy -q --workspace -- -D warnings 2>&1 && cargo nextest run --cargo-quiet --workspace --status-level fail 2>&1", goal_gate=true, retry_target="fixup"]
fixup [label="Fixup", prompt="The verify step failed. Read the build output from context and fix all clippy lint warnings and test failures.", max_visits=3]
fmt [label="Format", shape=parallelogram, script="cargo fmt --all 2>&1", max_retries=0]
start -> toolchain
toolchain -> preflight_compile [condition="outcome=success"]
toolchain -> exit
preflight_compile -> preflight_lint [condition="outcome=success"]
preflight_compile -> exit
preflight_lint -> implement [condition="outcome=success"]
preflight_lint -> fix_lints
fix_lints -> preflight_lint
implement -> simplify_opus -> simplify_gpt -> verify
verify -> fmt [condition="outcome=success"]
verify -> fixup
fixup -> verify
fmt -> exit
}
```
</details>
⚒️ Generated with [Fabro](https://fabro.sh)
---------
Co-authored-by: Fabro <noreply@fabro.sh>
|
||
|---|---|---|
| .. | ||
| fabro-agent | ||
| fabro-api | ||
| fabro-beastie | ||
| fabro-cli | ||
| fabro-config | ||
| fabro-db | ||
| fabro-devcontainer | ||
| fabro-git-storage | ||
| fabro-github | ||
| fabro-graphviz | ||
| fabro-hooks | ||
| fabro-interview | ||
| fabro-llm | ||
| fabro-mcp | ||
| fabro-model | ||
| fabro-openai-oauth | ||
| fabro-retro | ||
| fabro-sandbox | ||
| fabro-slack | ||
| fabro-telemetry | ||
| fabro-tracker | ||
| fabro-types | ||
| fabro-util | ||
| fabro-validate | ||
| fabro-workflows | ||