Across 17 comparable benchmarks, Inkling 2 Small has better reported results on 17 and worse on 0. 2 further shared results are listed below, each with the reason it is not counted.
Moderate evidence17 of 19 shared results countedHow we compare
Price basis: Standard rates. The price section below shows the applicable periods.
| Fact | Inkling 2 Small | Inkling 2 Mini |
|---|---|---|
| Context window | 1M tokens | 1M tokens |
| Input / 1M | $0.300 | $0.150 |
| Output / 1M | $1.00 | $0.500 |
| Blended (4:1) | $0.246 | $0.123 |
| Status | Available | Available |
| Released | November 5, 2026 | November 20, 2026 |
Blended is 4:1 input to output with 90% cache hits at listed rates — set your own mix.
Mean of 10 shared benchmarks grouped by capability on a 0–100 scale. How averages are built
Inkling 2 Small ahead on 2 of 2 counted
Inkling 2 Small ahead on 1 of 1 counted
Inkling 2 Small ahead on 3 of 3 counted
Inkling 2 Small ahead on 1 of 1 counted
No fixed-scale average — see the rows in Benchmark results
Inkling 2 Small ahead on 1 of 1 counted
Inkling 2 Small ahead on 4 of 4 counted
Inkling 2 Small ahead on 2 of 2 counted
No fixed-scale average — see the rows in Benchmark results
No shared comparable results for Mathematics, Science & health, Cybersecurity.
Published scores, with matching modes where available. Gaps show Inkling 2 Small relative to Inkling 2 Mini. Reading the results
| Benchmark | Inkling 2 Small | Inkling 2 Mini | Gap |
|---|---|---|---|
| DeepSWERepository Engineering · Pass@1 (DeepSWE v1.1) | 72.1Extra-High · tools | 48.2≠Thinking · tools | +23.9 pts#20 vs #61 of 73 |
| Vibe Code Bench v1.1Code Generation & Editing · Pass rate (%) | 81.6Extra-High · tools | 77.9≠Thinking · tools | +3.7 pts#30 vs #38 of 121 |
| ARC-AGI-1Abstract Generalization · Score (%) | 96.5Extra-High · no tools | 84.5≠Thinking · no tools | +12.0 pts#13 vs #47 of 87 |
| ARC-AGI-2Abstract Generalization · Score (% solved); cost per task (USD)Not counted · Speed and cost telemetry is reported, not scored | 81.7Extra-High · no tools | 49.3≠Thinking · no tools | 1.66×#21 vs #45 of 80 |
| AutomationBench-AATool Orchestration · Guardrail-safe objective completion rate | 45.6Extra-High · tools | 41.8≠Thinking · tools | +3.8 pts#16 vs #17 of 18 |
| MCP-AtlasTool Orchestration · Pass rate / claim coverage | 85.4Extra-High · tools | 82.7≠Thinking · tools | +2.7 pts#7 vs #14 of 55 |
| ScreenSpot-Pro (no tools)GUI Grounding · Success rate (%) | 83.8Extra-High · no tools | 76.5≠Thinking · no tools | +7.3 pts#2 vs #4 of 8 |
| DeepSearchQAFact Finding · F1 score | 90.9Extra-High · tools | 85.3≠Thinking · tools | 1.07×#2 vs #4 of 9 |
| LongBench v2Long Reasoning · Accuracy | 70.6Extra-High · no tools | 66.9≠Thinking · no tools | +3.7 pts#1 vs #2 of 22 |
| ChartographyDocuments & Charts · Score | 80.4Extra-High · tools | 72.5≠Thinking · tools | 1.11×#10 vs #16 of 29 |
| GDP.pdfDocuments & Charts · Task pass rate (%) | 25.2Extra-High · no tools | 16.8≠Thinking · no tools | +8.4 pts#17 vs #38 of 96 |
| MMMU-ProVisual Understanding · Accuracy | 81.6Extra-High · no tools | 74.9≠Thinking · no tools | +6.7 pts#25 vs #69 of 136 |
| PerceptionBenchVisual Understanding · Accuracy | 64.7Extra-High · no tools | 57.5≠Thinking · no tools | +7.2 pts#6 vs #11 of 19 |
| AA-BriefcaseOffice & Business · AA-Briefcase Elo | 1527Extra-High · tools | 1263≠Thinking · tools | 1.21×#15 vs #18 of 20 |
| GDPval-AA v2Cross-industry Work · Elo score | 1630Extra-High · tools | 1319≠Thinking · tools | 1.24×#14 vs #19 of 23 |
| JobBenchCross-industry Work · Mean task reward (%)Not counted · This metric has no better direction | 70.5Extra-High · tools | 49.6≠Thinking · tools | — |
| HLE-DiamondKnowledge Exams · Accuracy (%) | 28.7Extra-High · no tools | 15.2≠Thinking · no tools | +13.5 pts#9 vs #13 of 16 |
| AA Intelligence Index v5.0.1Capability Indices · Index (0-100) | 30.6Extra-High · tools | 20.1≠Thinking · no tools | 1.52×#30 vs #44 of 62 |
| Vals Index v2.2Capability Indices · GDP-weighted accuracy (%) | 56.1Extra-High · tools | 46.5≠Thinking · tools | +9.6 pts#21 vs #31 of 36 |
Repository Engineering · Pass@1 (DeepSWE v1.1)
Code Generation & Editing · Pass rate (%)
Tool Orchestration · Guardrail-safe objective completion rate
Tool Orchestration · Pass rate / claim coverage
GUI Grounding · Success rate (%)
Fact Finding · F1 score
Long Reasoning · Accuracy
Documents & Charts · Score
Documents & Charts · Task pass rate (%)
Visual Understanding · Accuracy
Visual Understanding · Accuracy
Office & Business · AA-Briefcase Elo
Cross-industry Work · Elo score
Cross-industry Work · Mean task reward (%)
Not counted · This metric has no better direction
Knowledge Exams · Accuracy (%)
Capability Indices · Index (0-100)
Capability Indices · GDP-weighted accuracy (%)
A dash means no result is available.
| Benchmark | Inkling 2 Small | Inkling 2 Mini |
|---|---|---|
| GPQA Diamond | 93.0 | — |
| HLE | 44.1 | — |
| HMMT Feb 2026 | 95.8 | — |
| MathArena Apex | 63.5 | — |
| Terminal-Bench-Science 0.1 | 5.7 | — |
| FrontierCode 2.0 | — | 15.1 |
| FrontierCS 2 | 16.4 | — |
| SWE-bench Multilingual | 81.2 | — |
| SWE-Bench Pro V2 | 97.7 | — |
| SWE-Bench Pro V2 Hard | 86.3 | — |
| SWE-bench Verified | 82.2 | — |
| SWE-Marathon | 38.8 | — |
| Terminal-Bench 2.1 | 85.1 | — |
| Terminal-Bench 4.0 | 13.1 | — |
| Terminal-Bench 4.1 | 13.0 | — |
| AA-LCR | 83.3 | — |
| OpenAI MRCR v2 (8-needle, 256K–512K) | — | 46.4 |
| BabyVision | 93.7 | — |
| CharXiv RQ | 87.7 | — |
| MathVision | 92.8 | — |
| OmniDocBench | — | 92.8 |
| ZeroBench Main | — | 34.0 |
| AA-AnalystAgent | — | 46.3 |
| Legal Research Bench | 46.6 | — |
| Tax Agent Bench | 65.8 | — |
| τ³-Banking | 40.4 | — |
| BrowseComp | — | 79.0 |
| AA-Omniscience | 16.1 | — |
| BioMysteryBench | 70.0 | — |
Time-of-day providers use peak-hour rates; fixed-price providers use standard rates. Off-peak discounts remain in the published details.
Drag the handle above the bar for cache hits, the one below it for the input : output mix · 5% steps
Blended price = effective input rate at 90% cache hits × 80% + output rate × 20%
Inkling 2 Small costs100%morethan Inkling 2 Mini per million tokens — 2× the price.Using standard rates at this token mix.
Inkling 2 Small: Standard pricing. Tinker 64K/256K inference; training billed separately. Image-specific token accounting not separately specified.
Inkling 2 Mini: Standard pricing. Tinker 64K/256K training (1M sampling); fine-tuning billed separately ($0.75/1M for 64K, $1.50/1M for 256K).
| Price / 1M tokens | Inkling 2 Small | Inkling 2 Mini | Ratio |
|---|---|---|---|
| Input | $0.30 | $0.15 | 2.00× |
| Output | $1.00 | $0.50 | 2.00× |
| Cached input | $0.03 | $0.015 | 2.00× |
| Cache write | — | — | — |
| Blended | $0.246 | $0.123 | 2.00× |
Estimates use the selected token mix. Cache writes, tool fees and long-context surcharges are excluded.
Reported cost per task, paired with the score from the same run. Settings and sources may differ; costs do not affect the verdict.
| Benchmark | Inkling 2 Small | Inkling 2 Mini |
|---|---|---|
| AA Intelligence Index v5.0.1 | Not reported | |
| AA-Briefcase | Not reported | |
| AA-Omniscience | Not reported | |
| ARC-AGI-1 | Not reported | |
| ARC-AGI-2 | Not reported | |
| AutomationBench-AA | $0.0581 / task Average · Guardrail-safe objective completion rate: 45.6 Extra-High (With Tools) | Not reported |
| FrontierCS 2 | Not reported | |
| GDP.pdf | Not reported | |
| HLE-Diamond | Not reported | |
| Terminal-Bench 4.0 | $8.22 / task Average · Resolution rate (%): 12.6 Unspecified Reasoning (With Tools) · vals-mini-swe | Not reported |
| Terminal-Bench 4.1 | Not reported | |
| Terminal-Bench-Science 0.1 | Not reported |
| Attribute | Inkling 2 Small | Inkling 2 Mini |
|---|---|---|
| Availability | Available | Available |
| Provider | Thinking Machines Lab | Thinking Machines Lab |
| Model type | Reasoning model | Reasoning model |
| First release | November 5, 2026 | November 20, 2026 |
| Context window | 1M tokens | 1M tokens |
| Total parameters | 594B | 26.4B |
| Active parameters | 17B | 5.4B |
| Architecture | Mixture of Experts (MoE) | Mixture of Experts (MoE) |
| Documented capabilities | Reasoning, Coding, Tool use, Long context | Reasoning, Coding, Tool use, Long context |
| Input modalities | Text, Image | Text, Image |
| Output modalities | Text | Text |
| Weights | Open weights | Open weights |
| Licence | Apache 2.0 | Apache 2.0 |
| Commercial use | Commercial use allowed | Commercial use allowed |
These are reported-score comparisons, not controlled head-to-head tests.
What people found comparing these models, or using either one. Comments also appear on the models' own pages.
Used Inkling 2 Small or Inkling 2 Mini? Sign in with Google to share what worked, what did not, and which API or setup you tested.
No community comments on either model yet.
Alternatives with the most benchmark overlap for this matchup.