gpt-5.5
Ranked 10 of 14 overall. Strongest on structured products at 93.1, weakest on credit analysis at 80.1.
Score
84.3
rank 10 of 14
Latency
87.2s
median per task
Cost
$0.817
per task at list prices
Capability profile
Mean score per category on question-answering, 0–100.
Score by category
Score and number of tasks per category, on question-answering.
Credit Analysis80.1 · 62
Monitoring91.9 · 20
Structured Products93.1 · 18
Research Intelligence81.5 · 10
Discovery & Prediction80.2 · 2
Where it sits
gpt-5.5 against every other model on the board.
gpt-5.5 beaten on both by kimi-k3 and 7 othersBest trade-off · no model beats these on bothBest-by-Task · hypothetical, not a model
On the research track
How it worked when it had to find its own evidence.
Delivered an answer
100%
of research tasks
Average steps
23.4
searches, reads and calculations per task
Where it lost the most
Its lowest-scoring tasks on question-answering.
PARTIAL
Coverage Metrics
Financial Foundations
PARTIAL
Fcf Debt Service
Financial Foundations
PARTIAL
Breakeven Analysis
Credit Analysis
PARTIAL
Trade Idea Extraction
Research Intelligence
PARTIAL
Spread Analysis
Credit Analysis