Expand description
2026-09-26: The serve matrix’s per-round probes: each turns its requests into
a Signal (or a tok/s figure); the bars that read them are in score.rs.
Owner: bench, serve matrix. Invariants:
- Every probe request is greedy (temperature 0) and carries no sampling
penalty (
base_body, andcoherence::probefor the coherence leg). A repetition penalty would penalise the repeated\nthatcode_shaperequires.
Structs§
- Coherence
- 2026-09-26: What one round’s coherence leg established.
Constants§
- NEEDLE
- 2026-09-26: Needle for the long-context probe; a test asserts the filler never contains it.
Functions§
- code_
shape - 2026-09-26: The structural test on the reply text.
- codegen_
probe - 2026-09-26: Structural codegen: does the reply define
fibwith an indented body (code_shape)? The reply is not executed. - coherence_
probe - 2026-09-26: Run
crate::coherence::probe(thecoherence::CHECKSquestions and the model-list check) and map it to the round’s coherence count and identity signal. The string is the probe’s concern, if any. - long_
context_ probe - 2026-09-26: Needle-in-a-haystack with a
tokens-token filler. - tool_
call_ probe - 2026-09-26: Did the model call
get_weatherfor Paris? A 4xx refusal of the request isNotApplicable(is_client_rejection); any other error, or a reply without that call, is aFail. - tps_
probe - 2026-09-26: Decode tok/s from the client’s TPOT (
ChatOutcome::tpot_ms) on abudget-token request.Nonewhen TPOT is undefined (no delta arrived, or fewer than two completion tokens); a failed request reports 0 with its error.