Kaelum Bench v0.2 · Public Calibration

Gemini 3.5 Flash

Accounting Knowledge · Run evidence

Provisional · Hybrid

Provisional calibration result — not a verified v1 result.

Configuration-specific evidence for a single published run. It is not a statistical ranking and is not an overall score.

Run identity

Model and configuration

Model display name
Gemini 3.5 Flash
Requested model ID
google/gemini-3.5-flash
Provider
Google Vertex AI
Reasoning configuration
high
Suite
Accounting Knowledge
Validity
Provisional

Published score

Scoring evidence

Score
541.6 / 600
Percentage
90.3%
Scoring method
Hybrid
Comparison protocol
Configuration-specific
Tool access
None
Published total cost (model + judge)
$1.63

Usage

Tokens and execution

Input tokens (tested model)
52,030
Output tokens (tested model)
80,454
Total tokens (tested model)
132,484
Attempted tasks
60 / 60
Valid responses
58 / 60
Invalid JSON
2
Schema failures
0
Provider failures
0
Missing responses
0

Release provenance

Versioned run evidence

Benchmark ID
FMB-KNOW-ACC-01
Benchmark version
0.1.0-dev
Scorer ID
fmb-live-scorer-v1
Started at
Completed at
Run ID
79e6d34b-f510-50de-92f3-9613526f0b63
Signed release ID
0bae13a5-a0c9-44fb-b6df-6bc83a6847d5