Launch report
Inkling 2 Small launch report
ModelCap · generated from recorded data
Thinking Machines's Inkling 2 Small was first listed on 9 November 2026. It entered the ModelCap Index at #91 of 359 ranked language models on 10 November 2026, one day after its listed release date, backed by 4 admitted benchmarks: ARC-AGI-2 Verified, Arena Coding, Arena Overall and LMArena Agent. Its first measurement, on 9 November 2026 on one benchmark, ARC-AGI-2 Verified, had placed it at #77 of 354; ModelCap counts a model as entered once two benchmarks back its position. As of 19 November 2026 it is #98 on the current board.
Position at entry
Entry is the first board position backed by at least two admitted benchmarks. Read from the board's last published snapshot of each day named below; these values are fixed and do not change with later refreshes.
- Listed release date
- 9 November 2026
- First measurement · 9 November 2026
- #77 of 354 ranked language modelsBacked by ARC-AGI-2 Verified
- Entry · 10 November 2026
- #91 of 359 ranked language modelsBacked by ARC-AGI-2 Verified, Arena Coding, Arena Overall, LMArena Agent
- Neighbours at entry
- Directly above: Qwen3 235B A22B Instruct 2507Directly below: DeepSeek V3.1 Terminus
Benchmark results
Published results admitted to the Index, as of 19 November 2026. Each board links to ModelCap's page for it and each result to its source.
Against Inkling Small
Inkling Small is the previous release in the same family. Current results for both, as of 19 November 2026.
- ModelCap Index
- Inkling 2 Small: #98 on the current board · score 56.8Inkling Small: #201 across all versions · score 52.0
How this report is produced
ModelCap publishes a launch report when a newly released model first holds a measured position backed by at least two public benchmarks. The report is assembled automatically from ModelCap's own records: the board as it stood at entry, the results currently admitted to the Index and, where one exists, the estimate recorded before measurement. No text is written by hand, and no claim is made that the records do not contain. ModelCap does not run the benchmarks itself; each result links to the organisation that published it.