Inkling 2 Small vs MiMo V2.6 Pro
Both models appear in the same evaluation table on the official model card, which is the only place they are measured under identical conditions and reported by the same party.
Across 43 benchmarks where both models report a score, Inkling 2 Small is higher on 30 and MiMo V2.6 Pro is higher on 12 , with 1 exact ties. Inkling 2 Small has 594B total parameters with 17B active, and is priced at $1 per million output tokens on Tinker. Its largest margin is on DesignArena Game Dev. Its largest deficit is on GDPVal-AA v2.1.
Head to head
- Position
- Inkling 2 Small: the subject of this site · MiMo V2.6 Pro: open weights, larger class
- Bechmarks scored
- 43 shared rows, out of 71 total
- Record
- 30 wins · 12 losses · 1 ties
Largest differences
| Benchmark | Inkling 2 Small | MiMo V2.6 Pro | Gap |
|---|---|---|---|
| DesignArena Game Dev · Visual Creation | 1,428 | 1,315 | + 113 |
| DesignArena 3D Design · Visual Creation | 1,401 | 1,333 | + 68 |
| GDPVal-AA v2.1 · Knowledge Work | 1,630 | 1,689 | -59 |
| AA-3D · Visual Creation | 1,021 | 977 | + 44 |
| AA-Omniscience v1.1 · Factuality | 73.3% | 39.8% | + 33.5 |
| DesignArena Website · Visual Creation | 1,340 | 1,307 | + 33 |
| Code2Video-AA · Visual Creation | 1,020 | 1,052 | -32 |
| Code Arena · Visual Creation | 1,653 | 1,625 | + 28 |
| ScreenSpot Pro · Visual Perception | 83.8% | 56.6% | + 27.2 |
| Chartography · Visual Reasoning | 36.7% | 10.8% | + 25.9 |
Every shared benchmark
| Benchmark | Inkling 2 Small | MiMo V2.6 Pro | Leader |
|---|---|---|---|
| Third Party Index | |||
| AA Index (v5.0) | 30.6 | 29.9 | Inkling 2 Small |
| Vals Index (v2.2) | 56.1% | 57.8% | MiMo V2.6 Pro |
| Coding | |||
| FrontierCode v2.0 | 25.4% | 24% | Inkling 2 Small |
| DeepSWE v1.1 | 72.1% | 71.9% | Inkling 2 Small |
| Terminal Bench Science v0.1 | 5.7% | 5.7% | tie |
| Terminal Bench v4.0 | 13.1% | 34.9% | MiMo V2.6 Pro |
| Terminal Bench 2.1 · Inkling 2 Small: Pi harness | 85.1% | 89.9% | MiMo V2.6 Pro |
| Vibe Code Bench v1.1 | 81.6% | 85.2% | MiMo V2.6 Pro |
| SWEBench Multimodal v2 | 43.1% | 41.3% | Inkling 2 Small |
| Search and Tools | |||
| MCP Atlas · Full / Public | 85.4% | 84.4% | Inkling 2 Small |
| Knowledge Work | |||
| AA-Briefcase v1.1 | 1,527 | 1,512 | Inkling 2 Small |
| GDPVal-AA v2.1 | 1,630 | 1,689 | MiMo V2.6 Pro |
| JobBench | 70.5% | 62% | Inkling 2 Small |
| Enterprise Operations | |||
| AutomationBench v1.0.6 | 45.6% | 58.6% | MiMo V2.6 Pro |
| Finance Work | |||
| Finance Agent v2.1 | 67.5% | 64.3% | Inkling 2 Small |
| Excel Modeling Benchmark v1.1 | 69.5% | 67.4% | Inkling 2 Small |
| Tax Agent Bench | 65.8% | 64.9% | Inkling 2 Small |
| Legal Work | |||
| Harvey LAB v1.1 | 21.7% | 13.8% | Inkling 2 Small |
| Legal Research Bench | 46.6% | 47.1% | MiMo V2.6 Pro |
| Visual Creation | |||
| DesignArena Web Apps | 1,296 | 1,275 | Inkling 2 Small |
| DesignArena Website | 1,340 | 1,307 | Inkling 2 Small |
| DesignArena Game Dev | 1,428 | 1,315 | Inkling 2 Small |
| DesignArena 3D Design | 1,401 | 1,333 | Inkling 2 Small |
| Code Arena | 1,653 | 1,625 | Inkling 2 Small |
| AA-3D | 1,021 | 977 | Inkling 2 Small |
| Code2Video-AA | 1,020 | 1,052 | MiMo V2.6 Pro |
| Visual Perception | |||
| ScreenSpot Pro · original / with Python | 83.8% | 56.6% | Inkling 2 Small |
| Humanity’s Sixth Sense | 33.3% | 25.2% | Inkling 2 Small |
| Visual Reasoning | |||
| GDP.pdf | 25.2% | 19.2% | Inkling 2 Small |
| MMMU Pro (Standard 10) | 81.6% | 74.1% | Inkling 2 Small |
| Charxiv RQ · original / with Python | 85.6% | 74.2% | Inkling 2 Small |
| Chartography · original / with Python | 36.7% | 10.8% | Inkling 2 Small |
| SAGE v1.1 | 88.6% | 84.4% | Inkling 2 Small |
| STEM Reasoning | |||
| CritPt v2 · Full / Hard | 55.7% | 56.9% | MiMo V2.6 Pro |
| HLE-Diamond | 28.7% | 15.2% | Inkling 2 Small |
| HLE (text only) | 44.1% | 49.4% | MiMo V2.6 Pro |
| GPQA Diamond | 93% | 89% | Inkling 2 Small |
| Long Context | |||
| AA-LCR v1.1 | 83.3% | 86.3% | MiMo V2.6 Pro |
| LongBench v2 | 70.6% | 68% | Inkling 2 Small |
| Factuality | |||
| AA-Omniscience v1.1 · score | 16.1 | -4.5 | Inkling 2 Small |
| AA-Omniscience v1.1 · non-hallucination rate | 73.3% | 39.8% | Inkling 2 Small |
| Instruction Following | |||
| ComplexConstraints v1.0 | 44.5% | 32.3% | Inkling 2 Small |
| Cybersecurity | |||
| CyberBench v1.1 | 72.7% | 72.9% | MiMo V2.6 Pro |
Reading the comparison honestly
These numbers come from one table published by Thinking Machines, using third-party evaluations where those exist. That makes them comparable to each other, but it does not make them neutral: the vendor chose which benchmarks to run and how to present them, and a benchmark that a model loses has a way of not appearing.
The models are also not the same shape of thing. MiMo V2.6 Pro sits in the open weights, larger class tier of the card's own grouping, while Inkling 2 Small is the subject of this site. A win against a smaller model is less meaningful than a win against a larger one, and the table does not adjust for that.