AmazonNova 2.0 Pro Preview
Nova 2.0 Pro Preview is an Amazon proprietary reasoning model, released 2 Dec 2025. Its best configuration (medium reasoning effort) ranks #369 of 785 on the BenchLeader Index at 50.5 ±7.9, in the upper half. The ± is the point: 438 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in instruction following (77) and lowest in knowledge (39). At $3.44 per million tokens blended it is among the most expensive ranked models. Output speed of 125 tokens per second puts it faster than most, with a first answer in 27.4 s. It has been measured at 4 reasoning-effort settings; this summary describes the best-scoring one, and the tabs above switch between them. Last measured 2 Dec 2025.stale: no new result in six months
- Blended price
- $3.44/M
- $1.25 in · $10.00 out
- Output speed
- 125 tok/s
- measured by Artificial Analysis
- First answer
- 27 s
- Context
- 256k
- Full answer
- 31 s
- median, reasoning included
- Cost per run
- –
- Released
- 2 Dec 2025
- Overall index51
- Reasoning41
- Coding44
- Agents & tools55
- Knowledge39
- Instruction following77
- Multimodal47
- Long context57
- Composite45
Reasoning-effort configurations
The same model behaves differently depending on how much it is allowed to think. Each row is one setting, scored only on the benchmarks that were run at that setting. “Not stated” collects results from publishers that did not say which setting they used; for a reasoning model that is usually its thinking mode, but we do not assume it. Pick a setting here or at the top of the page to see its price, speed and category scores.
| Effort | Index | Rank | Speed | First answer | Chat reply cost | Categories |
|---|---|---|---|---|---|---|
| no reasoning | 43.5 | #566 | 111 tok/s | 0.98 s | $0.0035 | Agents & tools 48 · Coding 24 · Composite 41 · Instruction following 53 · Knowledge 39 · Long context 39 · Reasoning 41 |
| low | 49.3 | #402 | 119 tok/s | 27 s | $0.0035 | Agents & tools 49 · Coding 38 · Composite 44 · Instruction following 77 · Knowledge 41 · Long context 57 · Multimodal 45 · Reasoning 41 |
| mediumbest | 50.5 | #369 | 125 tok/s | 27 s | $0.0035 | Agents & tools 55 · Coding 44 · Composite 45 · Instruction following 77 · Knowledge 39 · Long context 57 · Multimodal 47 · Reasoning 41 |
| not stated | – | – | – | – | – |
How its index has moved
9 Sept 2026 to 21 Oct 2026- medium
- low
- no reasoning
The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.
Benchmark results
See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | no reasoning | low | medium | Source | Trend |
|---|---|---|---|---|---|
| GPQA Diamond (AA)not in index | 63.6%#352 | 75.0%#252 | 78.5%#209 | Artificial Analysis | |
| Humanity's Last Exam (AA)not in index | 3.9%#556 | 5.2%#442 | 9.4%#335 | Artificial Analysis | |
| CritPt | 0.0%#315 | 0.0%#315 | 0.0%#315 | Artificial Analysis |
Coding
| Benchmark | no reasoning | low | medium | Source | Trend |
|---|---|---|---|---|---|
| SciCode | 28.1%#201 | 38.7%#173 | 42.7%#149 | SciCode | |
| Terminal-Bench 2.1 (AA)not in index | 17.2%#180 | 19.5%#176 | 29.6%#155 | Artificial Analysis |
Agents & tools
| Benchmark | no reasoning | low | medium | Source | Trend |
|---|---|---|---|---|---|
| Terminal-Bench Hard | 16.7%#184 | 17.4%#178 | 24.2%#144 | Artificial Analysis | |
| τ²-Bench Telecom (AA)not in index | 71.6%#151 | 90.6%#61 | 92.7%#51 | Artificial Analysis | |
| GDPval-AA v2.1not in index | 0.0%#124 | 0.0%#124 | 0.0%#124 | Artificial Analysis |
Knowledge
| Benchmark | no reasoning | low | medium | not stated | Source | Trend |
|---|---|---|---|---|---|---|
| AA-Omniscience | -49.5#386 | -46.5#361 | -49.0#383 | – | Artificial Analysis | |
| GDP.pdfnot in index | 2.0%#55 | – | – | 2.0%#55 | Epoch AI Benchmarking Hub | |
| AA-Omniscience: accuracynot in index | 16.9%#378 | 22.0%#278 | 22.0%#279 | – | Artificial Analysis | |
| AA-Omniscience: non-hallucinationnot in index | 20.1%#277 | 12.1%#381 | 9.0%#442 | – | Artificial Analysis | |
| Global-MMLU-Lite (AA)not in index | – | 87.6%#20 | – | – | Artificial Analysis |
Instruction following
| Benchmark | no reasoning | low | medium | Source | Trend |
|---|---|---|---|---|---|
| IFBench | 52.0%#163 | 79.6%#10 | 79.0%#12 | Artificial Analysis |
Multimodal
| Benchmark | low | medium | Source | Trend |
|---|---|---|---|---|
| MMMU-Pro | 62.7%#193 | 64.5%#184 | Artificial Analysis |
Long context
| Benchmark | no reasoning | low | medium | Source | Trend |
|---|---|---|---|---|---|
| AA-LCR | 30.0%#395 | 64.7%#255 | 64.0%#260 | Artificial Analysis |
Composite
| Benchmark | no reasoning | low | medium | Source | Trend |
|---|---|---|---|---|---|
| AA Intelligence Index v4.3.2 | 10.0#383 | 12.8#322 | 14.2#300 | Artificial Analysis |
Price history
The price above, per 1M tokens, every day it changed.
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0035 | – | 29.8 s |
| Summarize a 30-page report | 12,000 / 600 | $0.021 | – | 32.2 s |
| Code edit | 6,000 / 1,500 | $0.022 | – | 39.4 s |
| Agentic coding session | 60,000 / 4,000 | $0.115 | – | 59.4 s |
| Structured extraction | 2,000 / 200 | $0.0045 | – | 29.0 s |
See also
Questions people ask
- What are Nova 2.0 Pro Preview's benchmark scores?
- As of 25 Oct 2026, Nova 2.0 Pro Preview (medium effort) scores 50.5 ±7.9 on the BenchLeader Index, strongest in instruction following (77) and weakest in knowledge (39). That rests on 8 independent benchmark results, and every score on this page links to its publisher.
- How does Nova 2.0 Pro Preview rank?
- Nova 2.0 Pro Preview (medium effort) is #369 of 785 ranked configurations on 25 Oct 2026. 438 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Opus 5.5 (max effort) at 72.6.
- How much does Nova 2.0 Pro Preview cost per 1M tokens?
- As of 25 Oct 2026, Nova 2.0 Pro Preview costs $1.25 per 1M input tokens and $10.00 per 1M output tokens, $3.44 blended (three input tokens per output token). The list price last changed on 17 Oct 2026.
- When was Nova 2.0 Pro Preview released?
- Amazon released Nova 2.0 Pro Preview on 2 Dec 2025. BenchLeader has results for it at 4 reasoning-effort settings.
Data as of 25 Oct 2026. Compare these configurations.
Cite as: BenchLeader, “Nova 2.0 Pro Preview: benchmarks, pricing, speed and rank”, data as of 25 Oct 2026, https://www.benchleader.com/models/nova-2-0-pro.