AmazonNova Premier
Nova Premier is an Amazon proprietary model, released 30 Apr 2025. Its best configuration ranks #518 of 785 on the BenchLeader Index at 45.3 ±2.7, in the lower half. The ± is the point: 281 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 multimodal (53) and lowest in agents & tools (40). At $5.00 per million tokens blended it is among the most expensive ranked models. Last measured 24 Nov 2025.stale: no new result in six months
- Blended price
- $5.00/M
- $2.50 in · $12.50 out
- Output speed
- –
- First answer
- –
- Context
- 1M
- Full answer
- –
- Cost per run
- –
- Released
- 30 Apr 2025
- Overall index45
- Reasoning41
- Agents & tools40
- Maths47
- Knowledge46
- Instruction following45
- Multimodal53
- Long context41
- Composite40
How its index has moved
9 Sept 2026 to 21 Oct 2026The 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 | not stated | Source |
|---|---|---|
| GPQA Diamond (AA)not in index | 56.9%#398 | Artificial Analysis |
| Humanity's Last Exam (AA)not in index | 4.2%#531 | Artificial Analysis |
| Kagi LLM Benchmark | 44.8%#95 | Kagi LLM Benchmark |
| GPQA Diamond (HELM)not in index | 51.8%#36 | HELM Capabilities |
| CritPt | 0.0%#315 | Artificial Analysis |
Coding
| Benchmark | not stated | Source |
|---|---|---|
| SWE-bench Verified (any scaffold)not in index | 42.4%#42 | SWE-bench |
Agents & tools
| Benchmark | not stated | Source |
|---|---|---|
| Terminal-Bench Hard | 6.8%#249 | Artificial Analysis |
| τ²-Bench Telecom (AA)not in index | 38.3%#224 | Artificial Analysis |
Maths
| Benchmark | not stated | Source |
|---|---|---|
| Omni-MATH (HELM) | 35.0%#37 | HELM Capabilities |
Knowledge
| Benchmark | not stated | Source |
|---|---|---|
| AA-Omniscience | -35.4#302 | Artificial Analysis |
| MMLU-Pro (HELM)not in index | 72.6%#34 | HELM Capabilities |
| AA-Omniscience: accuracynot in index | 21.3%#286 | Artificial Analysis |
| AA-Omniscience: non-hallucinationnot in index | 28.1%#228 | Artificial Analysis |
Instruction following
| Benchmark | not stated | Source |
|---|---|---|
| IFBench | 36.2%#304 | Artificial Analysis |
| IFEval (HELM) | 80.3%#47 | HELM Capabilities |
| WildBench (HELM) | 78.8%#42 | HELM Capabilities |
Multimodal
| Benchmark | not stated | Source |
|---|---|---|
| MMMU (validation) | 68.0%#25 | MMMU |
| VTBnot in index | 2.0%#19 | Scale AI SEAL |
Long context
| Benchmark | not stated | Source |
|---|---|---|
| AA-LCR | 34.3%#379 | Artificial Analysis |
Composite
| Benchmark | not stated | Source |
|---|---|---|
| AA Intelligence Index v4.3.2 | 9.2#409 | Artificial Analysis |
| HELM Capabilities meannot in index | 63.7%#37 | HELM Capabilities |
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 at $0.625 per 1M. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0048 | $0.0042 | – |
| Summarize a 30-page report | 12,000 / 600 | $0.037 | $0.021 | – |
| Code edit | 6,000 / 1,500 | $0.034 | $0.025 | – |
| Agentic coding session | 60,000 / 4,000 | $0.200 | $0.116 | – |
| Structured extraction | 2,000 / 200 | $0.0075 | $0.0047 | – |
See also
Questions people ask
- What are Nova Premier's benchmark scores?
- As of 24 Oct 2026, Nova Premier scores 45.3 ±2.7 on the BenchLeader Index, strongest in multimodal (53) and weakest in agents & tools (40). That rests on 11 independent benchmark results, and every score on this page links to its publisher.
- How does Nova Premier rank?
- Nova Premier is #518 of 785 ranked configurations on 24 Oct 2026. 281 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 Premier cost per 1M tokens?
- As of 24 Oct 2026, Nova Premier costs $2.50 per 1M input tokens and $12.50 per 1M output tokens, $5.00 blended (three input tokens per output token).
- When was Nova Premier released?
- Amazon released Nova Premier on 30 Apr 2025.
Data as of 24 Oct 2026. Compare with another model.
Cite as: BenchLeader, “Nova Premier: benchmarks, pricing, speed and rank”, data as of 24 Oct 2026, https://www.benchleader.com/models/nova-premier.