fast-agent on Terminal-Bench 2.1, annotated.
The Terminal-Bench 2.1 runs of the fast-agent harness published on fast-agent.ai/benchmarks, reported trials, read with atif-scan as the leaderboard is in Terminal-Bench 2.1, annotated: what deserves a second look, what the traces leave out, and what each run cost. Most ran with a six-hour agent timeout, a modified evaluation that isn't comparable with leaderboard rows. Published scores as fast-agent.ai reports them.
Leaderboard
Our estimate: rewarded trials less review leads (an unexcused high or critical finding) and trials partly run by a fallback model. Strip: rewarded not counted in our estimate failed timed out errored no trajectory fallback model · corner: highest finding critical high medium
What the traces leave out
trials per row unless marked · shading is the share · click a figure for its trialsEach column is one kind of gap and how big it is. Output tokens: every output token, as tool calls messages reasoning with text in the trace, or generated and billed with no text in it (in practice reasoning), and why. Web content: search results and pages the model read whose text the trace doesn't hold, sized in input tokens from how much the prompt grew beyond what the trace shows. Compacted: earlier steps replaced by a summary. Our estimate of cost adds uncosted trials priced from their tokens and reasoning the output tokens undercount. How each figure is known (counted or inferred, fitted rates, undecided checks) is on the row's page.
Findings by family
medium and above · dot area: share of trials with an unexcused finding in the family · filled: some were rewarded · click for the row's findingsChecks × rows
Method
Scope
Each run's reported trials, from its public Harbor Hub jobs or its bench-run release in the published-benchmarks bucket, scanned with the inputs, flags and allowance rules fast-agent's docs use. Infrastructure failures rerun as replacements are left out, as the runs report them. Published is fast-agent.ai's score: rewarded trials over all 445, after fast-agent's own review, which disqualifies some rewarded trials. Our estimate applies this site's rule instead, as for every other row: rewarded trials less review leads and fallback-model trials. Costs are Harbor's recorded trial costs; on runs with fast-agent's accounting contract those are fast-agent's price-derived estimates (atif-scan won't call them a bill; they're shown as recorded, as fast-agent.ai does). .
Findings
Evidence that something happened in a trace, not a verdict; allowances excuse known-benign patterns per task. Review leads: rewarded trials with an unexcused high or critical finding.
Text per output token
Recorded agent text (reasoning, messages, tool arguments) over reported output tokens. ≈3.2 when the trace holds all output (measured on visible text over visible tokens where reasoning tokens are reported). In trace = that ratio over 3.2. With the visible text alone filling the reported tokens, recorded reasoning is missing from the token count and the cost (undercounted). Tokenizers differ: small deviations mean nothing.
Why output has no text in the trace
Source session files
Trace text
atif-scan's exports: credential-masked (best effort), literal text, nothing executed. Findings view for every trial; each trial links its Harbor Hub page or its files in the published-benchmarks bucket for the full trajectory.