GPT-5.6 Sol
Last updated: September 15, 2026
GPT-5.6 Sol completed 42 of 55 private-markets finance tasks (76.4%) at $2.76 per task and 90 seconds each, ranking #1 of 17 models tested — strongest on dashboards & widgets, weakest on error detection & repair.
Tested on 55 tasks across seven kinds of financial work — reading documents, looking up figures, building statements, finding and fixing errors, assembling dashboards, drawing charts, and multi-step requests. It passed 8 of 8 on dashboards & widgets and 4 of 7 on error detection & repair. Run-to-run noise is about ±1.8 tasks, so gaps of two tasks or fewer are ties.
Overall Rank
#1
of 17 engines tested
Pass Rate
76.4%
42 of 55 tasks passed
Judge Score
44.5
out of 55.0 available
Suite Split
37/49
analysis suite · 5/6 ingest harness
Cost per Task
$2.76
most expensive of 17
Cost, Full Pass
$152
all 55 tasks at list prices
Cost per Passed Task
$3.61
16th best value of 17
Latency
90s
per task · 1h 23m for a full pass
Category Breakdown
GPT-5.6 Sol against the average of every engine tested, category by category. The gap between the two bars is the only thing worth reading here — a category where the whole field struggles is a property of the tasks, not of the engine.
| Category | Passed | Pass Rate | Field Avg | Rank | Score | Cost/Task | Latency | Tool Calls |
|---|---|---|---|---|---|---|---|---|
| Document Ingestion | 5/6 | 83.3% | 42.2% | #1 of 17 | 5.90 | $11.61 | 364s | 79.7 |
| Metric Lookups | 12/12 | 100.0% | 88.7% | #1 of 17 | 12.00 | $0.44 | 28s | 7.2 |
| Financial Statements | 5/7 | 71.4% | 43.7% | #1 of 17 | 5.25 | $1.07 | 58s | 30.7 |
| Error Detection & Repair | 4/7 | 57.1% | 60.5% | #9 of 17 | 3.97 | $1.84 | 78s | 28.6 |
| Dashboards & Widgets | 8/8 | 100.0% | 84.6% | #1 of 17 | 7.92 | $3.30 | 99s | 39.0 |
| Charts | 3/4 | 75.0% | 66.2% | #1 of 17 | 3.02 | $1.78 | 68s | 17.8 |
| Multi-Step Requests | 4/6 | 66.7% | 54.9% | #1 of 17 | 4.70 | $2.97 | 47s | 21.3 |
Every column is measured within that category. The ingest harness is a separate suite of longer, heavier workflows — several times the tool calls of an analysis task — so its cost and latency are not comparable with the six rows below it.
Where it fell short
The 9 tasks GPT-5.6 Sol did not pass, of 50 graded individually. A benchmark that publishes only totals has to be taken on trust; these are the specific misses behind this model's score.
| Task | Kind of work | What it tests | Result |
|---|---|---|---|
| DD-01 | Document Ingestion | PDF → new rack, 5 components to the cent | Passed 1/2 0.90 runs |
| CTA-03 | Financial Statements | — | Failed · 0.15 |
| TRACE-01 | Financial Statements | — | Failed · 0.23 |
| DEN-01 | Error Detection & Repair | — | Failed |
| DEN-02 | Error Detection & Repair | — | Failed |
| DISC-02 | Error Detection & Repair | — | Failed · 0.03 |
| CHART-02 | Charts | — | Failed · 0.03 |
| CMP-03 | Multi-Step Requests | — | Failed · 0.45 |
| MT-03 | Multi-Step Requests | — | Failed · 0.25 |
A fraction means the task was run more than once and the runs disagreed. A number after a failure is the partial credit awarded. Five of the 55 tasks in the headline total are not broken out individually in the published results.
Against the Field
Every engine on the same axis, with GPT-5.6 Sol picked out. Grey bars are the rest of the field.
Overall Pass Rate
#1 of 17 · higher is better
Cost — One Full Pass (55 tasks)
#17 of 17 · lower is better
Cost per Passed Task
#16 of 17 · lower is better
Latency — Average per Task
#8 of 17 · lower is better
Tool Calls — Average per Task
#13 of 17 · lower is leaner
Cost vs Pass Rate
Up and to the left is better
Henon Tasks vs Tools benchmark — Ingest v5.1 + analysis suite v4, 17 executor models, 935 graded runs. Last updated September 15, 2026. Full results at henon.ai/benchmarks