Across 8 comparable benchmarks, Inkling 2 Mini has better reported results on 8 and worse on 0.
Moderate evidence8 of 8 shared results countedHow we compare
Price basis: Standard rates. The price section below shows the applicable periods.
| Fact | Inkling 2 Mini | Inkling |
|---|---|---|
| Context window | 1M tokens | 1M tokens |
| Input / 1M | $0.150 | $1.00 |
| Output / 1M | $0.500 | $4.05 |
| Blended (4:1) | $0.123 | $1.01 |
| Status | Available | Available |
| Released | November 20, 2026 | July 15, 2026 |
Blended is 4:1 input to output with 90% cache hits at listed rates — set your own mix.
Mean of 6 shared benchmarks grouped by capability on a 0–100 scale. How averages are built
Inkling 2 Mini ahead on 1 of 1 counted
Inkling 2 Mini ahead on 1 of 1 counted
Inkling 2 Mini ahead on 1 of 1 counted
Inkling 2 Mini ahead on 2 of 2 counted
Inkling 2 Mini ahead on 1 of 1 counted
No shared comparable results for Abstract reasoning, Mathematics, Science & health, Long context, Cybersecurity.
Published scores, with matching modes where available. Gaps show Inkling 2 Mini relative to Inkling. Reading the results
| Benchmark | Inkling 2 Mini | Inkling | Gap |
|---|---|---|---|
| Vibe Code Bench v1.1Code Generation & Editing · Pass rate (%) | 77.9Thinking · tools | 19.2≠Unspecified Reasoning · tools | +58.7 pts#38 vs #91 of 121 |
| MCP-AtlasTool Orchestration · Pass rate / claim coverage | 82.7Thinking · tools | 76.0≠Extra-High · tools | +6.7 pts#14 vs #28 of 55 |
| BrowseCompFact Finding · Accuracy | 79.0Thinking · tools | 77.1≠Thinking · tools + web | +1.9 pts#24 vs #29 of 61 |
| GDP.pdfDocuments & Charts · Task pass rate (%) | 16.8Thinking · no tools | 12.8≠Extra-High · no tools | +4.0 pts#38 vs #50 of 96 |
| MMMU-ProVisual Understanding · Accuracy | 74.9Thinking · no tools | 73.5≠Extra-High · no tools | +1.4 pts#69 vs #76 of 136 |
| AA-AnalystAgentData Analysis · pass^5 (%) | 46.3Thinking · tools | 23.8≠Thinking · tools + web | +22.5 pts#16 vs #27 of 39 |
| AA Intelligence Index v5.0.1Capability Indices · Index (0-100) | 20.1Thinking · no tools | 12.5≠Extra-High · tools | 1.61×#44 vs #51 of 62 |
| Vals Index v2.2Capability Indices · GDP-weighted accuracy (%) | 46.5Thinking · tools | 29.0≠Reported best | +17.5 pts#31 vs #35 of 36 |
Code Generation & Editing · Pass rate (%)
Tool Orchestration · Pass rate / claim coverage
Fact Finding · Accuracy
Data Analysis · pass^5 (%)
Capability Indices · Index (0-100)
Capability Indices · GDP-weighted accuracy (%)
A dash means no result is available.
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 Mini costs88%lessthan Inkling per million tokens — 8.2× cheaper.Using standard rates at this token mix.
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).
Inkling: Standard pricing. Thinking Machines Tinker official pricing (effective 2026-10-19) verified 2026-11-13; Inkling Serverless Inference Beta (256K)
| Price / 1M tokens | Inkling 2 Mini | Inkling | Ratio |
|---|---|---|---|
| Input | $0.15 | $1.00 | 0.15× |
| Output | $0.50 | $4.05 | 0.12× |
| Cached input | $0.015 | $0.17 | 0.09× |
| Cache write | — | — | — |
| Blended | $0.123 | $1.01 | 0.12× |
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 Mini | Inkling |
|---|---|---|
| AA Intelligence Index v5.0.1 | Not reported | |
| Code Migration | Not reported | $3.007 / task Average · Hidden-test pass rate (%): 11.8 Unspecified Reasoning (With Tools) |
| CWE-bench v0 | Not reported | |
| CWE-bench v1 | Not reported | |
| EMB (Excel Modeling) | Not reported | |
| Finance Agent v2 | Not reported | |
| FrontierSWE v2 | Not reported | |
| GDPx | Not reported | $6.70 / task Average · GDPx — Score: 2.3 Highest (source-reported) (With Tools) · overall/score |
| IOI (Vals v2) | Not reported | |
| Legal Research Bench | Not reported | |
| LiveBench 2.0 | Not reported | $0.9006 / task Average · Overall score (equal-weight mean of 7 suite scores, 0-100): 51.9 Max (task-specific tools) |
| MedCode | Not reported | |
| MedScribe | Not reported | |
| ProofBench v1.1 (Lean 4) | Not reported | |
| Public Benefits Bench v1.1 | Not reported | $0.3676 / task Average · Criterion pass rate (%): 58.5 Unspecified Reasoning (With Tools) |
| SAGE | Not reported | |
| Tax Agent Bench | Not reported | $0.3538 / task Average · Accuracy (%, rubric × citation quality): 15.3 Unspecified Reasoning (With Tools) |
| Terminal-Bench 4.0 | Not reported | $2.50 / task Average · Resolution rate (%): 0.51 Unspecified Reasoning (With Tools) · vals-mini-swe |
| Terminal-Bench 4.1 | Not reported | |
| Vals Index v2.1 | Not reported | $1.4644 / task Average · GDP-weighted accuracy (%): 28.7 Reported best (effort unspecified) |
| Vals Index v2.2 | Not reported | $1.8478 / task Average · GDP-weighted accuracy (%): 29.0 Reported best (effort unspecified) |
| Vibe Code Bench 1-100 | Not reported | |
| Vibe Code Bench v1.1 | Not reported | |
| WeirdML v3 | Not reported | $1.3283 / task Average · Official score (0-1; higher is better): 0.039 Extra-High (With Tools) |
| Attribute | Inkling 2 Mini | Inkling |
|---|---|---|
| Availability | Available | Available |
| Provider | Thinking Machines Lab | Thinking Machines Lab |
| Model type | Reasoning model | Multimodal model |
| First release | November 20, 2026 | July 15, 2026 |
| Context window | 1M tokens | 1M tokens |
| Total parameters | 26.4B | 975B |
| Active parameters | 5.4B | 41B |
| Architecture | Mixture of Experts (MoE) | Mixture of Experts (MoE) |
| Documented capabilities | Reasoning, Coding, Tool use, Long context | Reasoning, Multilingual, Coding, Tool use, Function calling, Long context |
| Input modalities | Text, Image | Text, Image, Audio |
| 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 Mini or Inkling? 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.