Model economics / detail
kimi-k2p6-turboC
Dispatch tier: Inexpensive (complexity class this model is routed for)
Twelve runs at 0.67 with a 29-second median: it sprints where its sibling k2p6 strolls to a perfect 54/54. Speed is charming, but a fast coin-flip-plus is still thin evidence. Come back with volume.
The measured record
What the ledger shows
Fleet-ledger aggregates for this model only — run counts, outcomes, spend, and token appetite. Missing telemetry reads “Not observed”, never zero.
Agent runs
12
Success rate
67%
Metered spend
Not observed
cost-blind: no run was metered
Cost / metered run
Not observed
Tokens in
Not observed
Tokens out
Not observed
Tokens in / run
Not observed
Tokens out / run
Not observed
Outcomes
Status breakdown
Every run ends in exactly one status. The bar is the whole record, to scale.
- failure: 4 of 12 runs (33.3% of total)
Run duration
How long the runs take
Wall-clock distribution across measured runs, from the fastest exit to the longest grind.
Min
2s
p25
6s
Median
29s
p75
2m
p90
3m
p99
6m
Max
7m
Mean
85s
Duration measured on 12 of 12 runs.
Commentary
Idiosyncrasies
What stands out in this model's numbers — shape, appetite, and failure habits.
- 29-second median — the fastest non-broken model in the fleet, with 10 of 12 runs done inside 3 minutes.
- 8-of-12 (67%): the turbo badge cost it the family’s trademark reliability (k2p6 ran 54-for-54).
- No cost or token telemetry on any run.
Commentary
Lessons learned
Practical routing and operations takeaways, grounded in the same record.
- Fine for jobs where a fast wrong answer is cheap to retry; 12 runs is not enough to say more.
- A fast coin-flip-plus is still a coin-flip-plus — watch the rate before routing it anything that matters.