Across 3 comparable benchmarks, Inkling 2 Mini has better reported results on 3 and worse on 0.
Moderate evidence3 of 3 shared results countedHow we compare
Price basis: Listed rates vs standard rates. The price section below shows the applicable periods.
| Fact | Nemotron 3.5 Bolt 74B-A9B | Inkling 2 Mini |
|---|---|---|
| Context window | 256K tokens | 1M tokens |
| Input / 1M | — | $0.150 |
| Output / 1M | — | $0.500 |
| Blended (4:1) | — | $0.123 |
| Status | Available | Available |
| Released | October 15, 2026 | November 20, 2026 |
Blended is 4:1 input to output with 90% cache hits at listed rates — set your own mix.
Mean of 2 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
No fixed-scale average — see the rows in Benchmark results
No shared comparable results for Coding, Abstract reasoning, Mathematics, Science & health, Long context, Multimodal, Cybersecurity.
Published scores, with matching modes where available. Gaps show Nemotron 3.5 Bolt 74B-A9B relative to Inkling 2 Mini. Reading the results
| Benchmark | Nemotron 3.5 Bolt 74B-A9B | Inkling 2 Mini | Gap |
|---|---|---|---|
| MCP-AtlasTool Orchestration · Pass rate / claim coverage | 55.6Unspecified Reasoning · tools | 82.7≠Thinking · tools | −27.1 pts#50 vs #14 of 55 |
| BrowseCompFact Finding · Accuracy | 70.0Unspecified Reasoning · tools | 79.0≠Thinking · tools | −9.0 pts#36 vs #24 of 61 |
| GDPval-AA v2Cross-industry Work · Elo score | 31.1Unspecified Reasoning · tools | 1319≠Thinking · tools | 0.02×#23 vs #19 of 23 |
Tool Orchestration · Pass rate / claim coverage
Fact Finding · Accuracy
Cross-industry Work · Elo score
A dash means no result is available.
| Benchmark | Nemotron 3.5 Bolt 74B-A9B | Inkling 2 Mini |
|---|---|---|
| GPQA Diamond | 88.8 | — |
| HLE | 28.7 | — |
| HLE-Diamond | — | 15.2 |
| DeepSWE | — | 48.2 |
| FrontierCode 2.0 | — | 15.1 |
| Terminal-Bench 2.1 | 53.9 | — |
| Vibe Code Bench v1.1 | — | 77.9 |
| AutomationBench-AA | — | 41.8 |
| PinchBench v2 | 90.3 | — |
| ScreenSpot-Pro (no tools) | — | 76.5 |
| AA Intelligence Index v5.0.1 | — | 20.1 |
| Vals Index v2.2 | — | 46.5 |
| AA-LCR | 67.4 | — |
| LongBench v2 | — | 66.9 |
| OpenAI MRCR v2 (8-needle, 256K–512K) | — | 46.4 |
| Chartography | — | 72.5 |
| GDP.pdf | — | 16.8 |
| MMMU-Pro | — | 74.9 |
| OmniDocBench | — | 92.8 |
| PerceptionBench | — | 57.5 |
| ZeroBench Main | — | 34.0 |
| AA-AnalystAgent | — | 46.3 |
| AA-Briefcase | — | 1263 |
| JobBench | — | 49.6 |
| τ³-Banking | 17.0 | — |
| ARC-AGI-1 | — | 84.5 |
| ARC-AGI-2 | — | 49.3 |
| 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%
A blended cost needs a published input and output price on both sides, and one of these two does not have them yet.
Nemotron 3.5 Bolt 74B-A9B: Listed text rates; pricing tier not confirmed.
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 | Nemotron 3.5 Bolt 74B-A9B | Inkling 2 Mini | Ratio |
|---|---|---|---|
| Input | — | $0.15 | — |
| Output | — | $0.50 | — |
| Cached input | — | $0.015 | — |
| Cache write | — | — | — |
| Blended | — | $0.123 | — |
Estimates use the selected token mix. Cache writes, tool fees and long-context surcharges are excluded.
| Attribute | Nemotron 3.5 Bolt 74B-A9B | Inkling 2 Mini |
|---|---|---|
| Availability | Available | Available |
| Provider | NVIDIA | Thinking Machines Lab |
| Model type | Multimodal model | Reasoning model |
| First release | October 15, 2026 | November 20, 2026 |
| Context window | 256K tokens | 1M tokens |
| Total parameters | 74B | 26.4B |
| Active parameters | 9B | 5.4B |
| Architecture | Mixture of Experts (MoE) | Mixture of Experts (MoE) |
| Documented capabilities | Reasoning, Multilingual, Tool use, Function calling, Structured output, Coding | Reasoning, Coding, Tool use, Long context |
| Knowledge cutoff | 2026-05 | — |
| Input modalities | Text, Image, Video | Text, Image |
| Output modalities | Text | Text |
| Weights | Open weights | Open weights |
| Licence | OpenMDW License Agreement 1.1 | 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 Nemotron 3.5 Bolt 74B-A9B 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.