POSIX → wasm32-wasi
The host the code assumes changes. 358,006 lines, 10 hours per model, and 2,668 checks across 7 modules.
- Runs
- 26
- Past rung 1
- 15
- Past rung 2
- 1
- Past rung 3
- 0
- Best composite
- 60
one per configuration
of 26 — the migration happened
of those 15 — every check perfect
of the 1 — accepted
highest score on this task
The instrument
Acceptance gates
6
decide whether the migration happened
Behavioral modules
7
ported suites, run against the new tree
Frozen checks
2,668
all must pass to open rung three
Verifiers
6
hostile programs, written against the spec
Where the 26 runs stopped
- 11
- Never migrated
- 14
- Migrated, behavior lost
- 0
- Perfect suite, gate rejected
- 1
- Verifier found a difference
- 0
- Accepted
1 run was perfect on all 2,668 checks, and passed the gate too. Of the 1 that reached rung three, 0 survived all six verifiers.
By model
| Model | Runs | Gate | Ceil. | Where they stopped | Mean | Best |
|---|---|---|---|---|---|---|
| claude-opus-5 | 5 | 5 | 1 | | 12.0 | 60 |
| claude-sonnet-5 | 5 | 2 | 0 | | 0.0 | — |
| dsv4-flash | 1 | 1 | 0 | | 0.0 | — |
| glm-5.2 | 1 | 0 | 0 | | 0.0 | — |
| gpt-5.6-luna | 6 | 3 | 0 | | 0.0 | — |
| gpt-5.6-sol | 6 | 3 | 0 | | 0.0 | — |
| kimi-k3 | 1 | 1 | 0 | | 0.0 | — |
| qwen3.8-max | 1 | 0 | 0 | | 0.0 | — |
Every run
best composite firstOne row per configuration. Shape is the run's tool sequence in 24 slices — pale is shell, dark is an edit — and each log is the full session as recorded.
| Model | Effort | Outcome | Gate | Checks | Verif. | Score | Shape |
|---|---|---|---|---|---|---|---|
| claude-opus-5 | xhigh | broken | 6/6 | 100.0% | 2/6 | 60 | |
| claude-sonnet-5 | xhigh | failed | 5/6 | 99.9% | — | 0 | |
| kimi-k3 | max | partial | 6/6 | 99.9% | — | 0 | |
| claude-sonnet-5 | max | partial | 6/6 | 99.9% | — | 0 | |
| claude-opus-5 | max | partial | 6/6 | 99.9% | — | 0 | |
| claude-opus-5 | low | partial | 6/6 | 99.8% | — | 0 | |
| qwen3.8-max | max | failed | 3/6 | 99.7% | — | 0 | |
| claude-opus-5 | medium | partial | 6/6 | 99.7% | — | 0 | |
| gpt-5.6-luna | max | failed | 4/6 | 99.7% | — | 0 | |
| gpt-5.6-sol | xhigh | partial | 6/6 | 99.6% | — | 0 | |
| claude-opus-5 | high | partial | 6/6 | 99.5% | — | 0 | |
| gpt-5.6-sol | max | failed | 5/6 | 99.2% | — | 0 | |
| dsv4-flash | max | partial | 6/6 | 99.0% | — | 0 | |
| glm-5.2 | max | failed | 5/6 | 99.0% | — | 0 | |
| gpt-5.6-luna | high | partial | 6/6 | 99.0% | — | 0 | |
| gpt-5.6-sol | none | partial | 6/6 | 99.0% | — | 0 | |
| gpt-5.6-sol | medium | partial | 6/6 | 99.0% | — | 0 | |
| gpt-5.6-luna | none | failed | 5/6 | 99.0% | — | 0 | |
| gpt-5.6-sol | high | failed | 5/6 | 99.0% | — | 0 | |
| gpt-5.6-luna | medium | partial | 6/6 | 98.9% | — | 0 | |
| claude-sonnet-5 | low | failed | 5/6 | 98.9% | — | 0 | |
| claude-sonnet-5 | medium | failed | 5/6 | 98.8% | — | 0 | |
| claude-sonnet-5 | high | partial | 6/6 | 98.8% | — | 0 | |
| gpt-5.6-luna | xhigh | partial | 6/6 | 98.7% | — | 0 | |
| gpt-5.6-luna | low | failed | 5/6 | 98.3% | — | 0 | |
| gpt-5.6-sol | low | failed | 4/6 | 87.8% | — | 0 |