Model detail
Inkling
Provider: Other · Public slug: thinkingmachines-inkling
- Rank
- Partial average — 2/3
- Rank movement
- Baseline
- Score
- 44.5
- Last measured
- 2026-07-26
- Coverage
- Missing Intensive
Reasoning profile
Available reasoning settings
NoneMinimalLowMediumHighMax
Checked on 2026-07-24 · provider source
Tested reasoning settings
| Provider setting | Score | Evidence | Filter bracket |
|---|---|---|---|
| None | 36.9 | Retained request evidence; eligible | Baseline |
| Medium | 52.1 | Retained request evidence; eligible | Balanced |
| Max | 33.9 | Retained request evidence; eligible | Intensive |
Filter assignment status
- Baseline: None — tested and eligible
- Balanced: Medium — tested and eligible
- Intensive: Max — tested and eligible
Generation health
Medium: 0% / 69% / 31% · n 16None: 44% / 31% / 25% · n 16 Missing IntensiveHigh repair rateHigh failure rateCost summary
- Mean standard list-price equivalent
- $0.09
- Median unique-cell cost
- $0.06
- Unique-cell range
- $0.01–$0.21
- Coverage
- 24 unique cells / 48 observed runs
- Observed-run total
- $4.17
- Evidence basis
- Provider-reported
- Mean output
- 18k tok
- Output range
- 1.4k–55k tok
- Token samples
- 48
Cost means equal-weight unique model/game/reasoning cells.
Run setup
- Overall Models: combines public reasoning variants
- Standard GameBench code-generation prompt