Inkling 2 Small and Phi-4-mini-reasoning have no benchmarks in common in this dataset, so their scores cannot be compared directly. The specification and pricing rows below still apply.
Price basis: Standard rates. The price section below shows the applicable periods.
| Fact | Inkling 2 Small | Phi-4-mini-reasoning |
|---|---|---|
| Context window | 1M tokens | 128K tokens |
| Input / 1M | $0.300 | $0.075 |
| Output / 1M | $1.00 | $0.300 |
| Blended (4:1) | $0.246 | $0.120 |
| Status | Available | Available |
| Released | November 5, 2026 | April 30, 2025 |
Blended is 4:1 input to output with 90% cache hits at listed rates — set your own mix.
Published scores, with matching modes where available. Gaps show Inkling 2 Small relative to Phi-4-mini-reasoning. Reading the results
A dash means no result is available.
| Benchmark | Inkling 2 Small | Phi-4-mini-reasoning |
|---|---|---|
| GPQA Diamond | 93.0 | — |
| HLE | 44.1 | — |
| HLE-Diamond | 28.7 | — |
| HMMT Feb 2026 | 95.8 | — |
| MathArena Apex | 63.5 | — |
| Terminal-Bench-Science 0.1 | 5.7 | — |
| DeepSWE | 72.1 | — |
| 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 | — |
| Vibe Code Bench v1.1 | 81.6 | — |
| AutomationBench-AA | 45.6 | — |
| MCP-Atlas | 85.4 | — |
| ScreenSpot-Pro (no tools) | 83.8 | — |
| AA Intelligence Index v5.0.1 | 30.6 | — |
| Vals Index v2.2 | 56.1 | — |
| AA-LCR | 83.3 | — |
| LongBench v2 | 70.6 | — |
| BabyVision | 93.7 | — |
| Chartography | 80.4 | — |
| CharXiv RQ | 87.7 | — |
| GDP.pdf | 25.2 | — |
| MathVision | 92.8 | — |
| MMMU-Pro | 81.6 | — |
| PerceptionBench | 64.7 | — |
| AA-Briefcase | 1527 | — |
| GDPval-AA v2 | 1630 | — |
| JobBench | 70.5 | — |
| Legal Research Bench | 46.6 | — |
| Tax Agent Bench | 65.8 | — |
| τ³-Banking | 40.4 | — |
| ARC-AGI-1 | 96.5 | — |
| ARC-AGI-2 | 81.7 | — |
| DeepSearchQA | 90.9 | — |
| 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 costs105%morethan Phi-4-mini-reasoning 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.
Phi-4-mini-reasoning: Standard pricing. Azure AI Foundry (Phi-4-mini-reasoning (128K)) verified 2026-11-14
| Price / 1M tokens | Inkling 2 Small | Phi-4-mini-reasoning | Ratio |
|---|---|---|---|
| Input | $0.30 | $0.075 | 4.00× |
| Output | $1.00 | $0.30 | 3.33× |
| Cached input | $0.03 | — | — |
| Cache write | — | — | — |
| Blended | $0.246 | $0.12 | 2.05× |
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 | Phi-4-mini-reasoning |
|---|---|---|
| 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 | Phi-4-mini-reasoning |
|---|---|---|
| Availability | Available | Available |
| Provider | Thinking Machines Lab | Microsoft |
| Model type | Reasoning model | Reasoning model |
| First release | November 5, 2026 | April 30, 2025 |
| Context window | 1M tokens | 128K tokens |
| Max output | — | 32K tokens |
| Total parameters | 594B | 3.8B |
| Active parameters | 17B | — |
| Architecture | Mixture of Experts (MoE) | Dense |
| Documented capabilities | Reasoning, Coding, Tool use, Long context | Reasoning, Multilingual |
| Input modalities | Text, Image | Text |
| Output modalities | Text | Text |
| Weights | Open weights | Open weights |
| Licence | Apache 2.0 | MIT License |
| 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 Phi-4-mini-reasoning? 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.