Across 3 comparable benchmarks, Amazon Nova Pro has better reported results on 2 and worse on 1.
Moderate evidence3 of 3 shared results countedHow we compare
Price basis: Standard rates vs listed rates. The price section below shows the applicable periods.
| Fact | Amazon Nova Pro | Llama3.1-70B-Instruct |
|---|---|---|
| Context window | 300K tokens | 128K tokens |
| Input / 1M | $0.800 | $0.720 |
| Output / 1M | $3.20 | $0.720 |
| Blended (4:1) | $1.28 | $0.720 |
| Status | Available | Available |
| Released | December 3, 2024 | July 23, 2024 |
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
Amazon Nova Pro ahead on 1 of 1 counted
Amazon Nova Pro ahead on 1 of 1 counted
No shared comparable results for Abstract reasoning, Science & health, Computer & tool use, Research & search, Long context, Multimodal, Professional work & productivity, Cybersecurity.
Published scores, with matching modes where available. Gaps show Amazon Nova Pro relative to Llama3.1-70B-Instruct. Reading the results
| Benchmark | Amazon Nova Pro | Llama3.1-70B-Instruct | Gap |
|---|---|---|---|
| HumanEvalCode Generation & Editing · Accuracy | 89.0Normal · no tools | 80.5Normal · no tools | +8.5 pts#18 vs #41 of 100 |
| MATHMathematics · Accuracy | 76.6Normal · no tools | 67.8Normal · no tools | +8.8 pts#14 vs #26 of 42 |
| MMLUKnowledge Exams · Accuracy | 85.9Normal · no tools | 86.0Normal · no tools | −0.1 pts#31 vs #29 of 114 |
A dash means no result is available.
| Benchmark | Amazon Nova Pro | Llama3.1-70B-Instruct |
|---|---|---|
| GPQA Diamond | — | 48.0 |
| HLE | 4.4 | — |
| MMLU-Pro | — | 66.4 |
| LiveCodeBench | — | 33.3 |
| MBPP | — | 86.0 |
| Terminal Bench Hard | 6.1 | — |
| WeirdML v2 | — | 9.0 |
| PinchBench v2 | — | 10.7 |
| ECI | 124 | — |
| Creative Writing | — | 782 |
| MMMU | 62.0 | — |
| MMMU-Pro | 44.3 | — |
| BALROG | — | 27.9 |
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%
Amazon Nova Pro costs78%morethan Llama3.1-70B-Instruct per million tokens.Using standard rates vs listed rates at this token mix.
Amazon Nova Pro: Standard pricing. AWS Bedrock official pricing (Amazon Nova Pro; US East N. Virginia) verified 2026-11-14
Llama3.1-70B-Instruct: Listed text rates; pricing tier not confirmed.
| Price / 1M tokens | Amazon Nova Pro | Llama3.1-70B-Instruct | Ratio |
|---|---|---|---|
| Input | $0.80 | $0.72 | 1.11× |
| Output | $3.20 | $0.72 | 4.44× |
| Cached input | — | — | — |
| Cache write | — | — | — |
| Blended | $1.28 | $0.72 | 1.78× |
Estimates use the selected token mix. Cache writes, tool fees and long-context surcharges are excluded.
| Attribute | Amazon Nova Pro | Llama3.1-70B-Instruct |
|---|---|---|
| Availability | Available | Available |
| Provider | Amazon | Facebook AI Research Lab |
| Model type | Multimodal model | Foundation model |
| First release | December 3, 2024 | July 23, 2024 |
| Context window | 300K tokens | 128K tokens |
| Total parameters | — | 70B |
| Architecture | Undisclosed | Dense |
| Documented capabilities | Multilingual | — |
| Input modalities | — | Text |
| Output modalities | — | Text |
| Weights | Closed / API only | Open weights |
| Licence | — | LLAMA 3.1 COMMUNITY LICENSE AGREEMENT |
| Commercial use | Proprietary | 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 Amazon Nova Pro or Llama3.1-70B-Instruct? 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.