Across 1 comparable benchmark, Inkling 2 Mini has better reported results on 1 and worse on 0.
Limited evidence1 of 1 shared result countedHow we compare
Price basis: Standard rates. The price section below shows the applicable periods.
| Fact | Qwen3-32B | Inkling 2 Mini |
|---|---|---|
| Context window | 128K tokens | 1M tokens |
| Input / 1M | $0.080 | $0.150 |
| Output / 1M | $0.280 | $0.500 |
| Blended (4:1) | $0.120 | $0.123 |
| Status | Deprecated | Available |
| Released | April 28, 2025 | November 20, 2026 |
Blended is 4:1 input to output with 90% cache hits at listed rates — set your own mix.
Mean of 1 shared benchmark grouped by capability on a 0–100 scale. How averages are built
Inkling 2 Mini ahead on 1 of 1 counted
No shared comparable results for Coding, Abstract reasoning, Mathematics, Science & health, Computer & tool use, Research & search, Long context, Professional work & productivity, Cybersecurity.
Published scores, with matching modes where available. Gaps show Qwen3-32B relative to Inkling 2 Mini. Reading the results
| Benchmark | Qwen3-32B | Inkling 2 Mini | Gap |
|---|---|---|---|
| GDP.pdfDocuments & Charts · Task pass rate (%) | 1.6Thinking · no tools | 16.8Thinking · no tools | −15.2 pts#86 vs #38 of 96 |
Documents & Charts · Task pass rate (%)
A dash means no result is available.
| Benchmark | Qwen3-32B | Inkling 2 Mini |
|---|---|---|
| AIME 2024 | 81.4 | — |
| AIME2025 | 20.2 | — |
| C-Eval | 87.3 | — |
| CritPt | 0.3 | — |
| GPQA Diamond | 68.4 | — |
| HLE-Diamond | — | 15.2 |
| MATH-500 | 97.2 | — |
| SciCode | 36.0 | — |
| Aider-Polyglot | 40.0 | — |
| CodeForces | 1977 | — |
| DeepSWE | — | 48.2 |
| FrontierCode 2.0 | — | 15.1 |
| LiveCodeBench | 65.7 | — |
| Terminal Bench Hard | 3.0 | — |
| Vibe Code Bench v1.1 | — | 77.9 |
| AutomationBench-AA | — | 41.8 |
| MCP-Atlas | — | 82.7 |
| ScreenSpot-Pro (no tools) | — | 76.5 |
| AA Intelligence Index v5.0.1 | — | 20.1 |
| ECI | 139 | — |
| LiveBench | 43.6 | — |
| Vals Index v2.2 | — | 46.5 |
| LongBench v2 | — | 66.9 |
| OpenAI MRCR v2 (8-needle, 256K–512K) | — | 46.4 |
| Chartography | — | 72.5 |
| MMMU-Pro | — | 74.9 |
| OmniDocBench | — | 92.8 |
| PerceptionBench | — | 57.5 |
| ZeroBench Main | — | 34.0 |
| AA-AnalystAgent | — | 46.3 |
| AA-Briefcase | — | 1263 |
| GDPval-AA v2 | — | 1319 |
| JobBench | — | 49.6 |
| τ³-Banking | 5.4 | — |
| ARC-AGI-1 | — | 84.5 |
| ARC-AGI-2 | — | 49.3 |
| BrowseComp | — | 79.0 |
| DeepSearchQA | — | 85.3 |
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%
Qwen3-32B costs2.3%lessthan Inkling 2 Mini per million tokens.Using standard rates at this token mix.
Qwen3-32B: Standard pricing. DeepInfra official models API (Qwen/Qwen3-32B) verified 2026-11-13
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 | Qwen3-32B | Inkling 2 Mini | Ratio |
|---|---|---|---|
| Input | $0.08 | $0.15 | 0.53× |
| Output | $0.28 | $0.50 | 0.56× |
| Cached input | — | $0.015 | — |
| Cache write | — | — | — |
| Blended | $0.12 | $0.123 | 0.98× |
Estimates use the selected token mix. Cache writes, tool fees and long-context surcharges are excluded.
| Attribute | Qwen3-32B | Inkling 2 Mini |
|---|---|---|
| Availability | Deprecated | Available |
| Provider | Alibaba | Thinking Machines Lab |
| Model type | Reasoning model | Reasoning model |
| First release | April 28, 2025 | November 20, 2026 |
| Context window | 128K tokens | 1M tokens |
| Max output | 16K tokens | — |
| Total parameters | 32B | 26.4B |
| Active parameters | 32B | 5.4B |
| Architecture | Dense | Mixture of Experts (MoE) |
| Documented capabilities | Reasoning, Multilingual | Reasoning, Coding, Tool use, Long context |
| Input modalities | Text | 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 Qwen3-32B 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.