Model comparison
Amazon Nova Pro vs DeepSeek-R1-Distill-Llama-70B
DeepSeek-R1-Distill-Llama-70B is the stronger model overall, scoring 37.4 to 30.5 on the Noometry Index.
Last verified . 7 shared benchmarks.
DeepSeekSummary
- They share 7 benchmarks with published results for both. Amazon Nova Pro scores higher in 0 categories and DeepSeek-R1-Distill-Llama-70B in 6 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek-R1-Distill-Llama-70B leads 34.3 to 26.4.
- The biggest single-benchmark swing is LiveBench Reasoning: 32.6% for Amazon Nova Pro and 67.6% for DeepSeek-R1-Distill-Llama-70B.
- DeepSeek-R1-Distill-Llama-70B has downloadable open weights; the other is API-only.
Side by side
| Amazon Nova Pro | DeepSeek-R1-Distill-Llama-70B | |
|---|---|---|
| Provider | Amazon | DeepSeek |
| Noometry Index | 30.5 | 37.4 |
| Released | 2024-12-03 | 2025-01-20 |
| Weights | Proprietary | Open |
| Context window | 300K | — |
| Max output | 10K | — |
| Input $ / M tokens | $0.80 | — |
| Output $ / M tokens | $3.20 | — |
| Results tracked | 38 | 13 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-R1-Distill-Llama-70B leads
Amazon Nova Pro: 34.6 (#224), DeepSeek-R1-Distill-Llama-70B: 36.3 (#197)
| Benchmark | Amazon Nova Pro | DeepSeek-R1-Distill-Llama-70B |
|---|---|---|
| LiveBench Coding | 38.1% | 51.6% |
| BigCodeBench Instruct | — | 35.3% |
| LMArena Coding | 1270 | — |
| BigCodeBench Complete | — | 49.9% |
Agentic & Tool Use Not comparable
Amazon Nova Pro: 16.6 (#139), DeepSeek-R1-Distill-Llama-70B: —
| Benchmark | Amazon Nova Pro | DeepSeek-R1-Distill-Llama-70B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 25% | — |
| TheAgentCompany | 1.7% | — |
Reasoning DeepSeek-R1-Distill-Llama-70B leads
Amazon Nova Pro: 19.2 (#240), DeepSeek-R1-Distill-Llama-70B: 24.1 (#150)
| Benchmark | Amazon Nova Pro | DeepSeek-R1-Distill-Llama-70B |
|---|---|---|
| LiveBench Reasoning | 32.6% | 67.6% |
| LiveBench Data Analysis | 48.3% | 55.9% |
| LiveBench | 43.5% | 54.5% |
| Kagi LLM Benchmark | — | 52.3% |
| LMArena Hard Prompts | 1246 | — |
| Epoch Capabilities Index | 123.75 | — |
Math DeepSeek-R1-Distill-Llama-70B leads
Amazon Nova Pro: 26.4 (#240), DeepSeek-R1-Distill-Llama-70B: 34.3 (#161)
| Benchmark | Amazon Nova Pro | DeepSeek-R1-Distill-Llama-70B |
|---|---|---|
| LiveBench Math | 38% | 58.1% |
| OTIS Mock AIME 2024-2025 | — | 51.4% |
| Omni-MATH | 24.2% | — |
| LMArena Math | 1252 | — |
| MATH Level 5 | — | 89.9% |
Knowledge DeepSeek-R1-Distill-Llama-70B leads
Amazon Nova Pro: 27.1 (#246), DeepSeek-R1-Distill-Llama-70B: 30.7 (#216)
| Benchmark | Amazon Nova Pro | DeepSeek-R1-Distill-Llama-70B |
|---|---|---|
| GPQA Diamond | — | 55.7% |
| Humanity's Last Exam | 4.4% | — |
| MMLU-Pro | 67.3% | — |
| Confabulations | 30.1% | — |
| Vectara Hallucination Rate | 5.1% | — |
| GPQA (HELM) | 44.6% | — |
| LMArena Expert | 1211 | — |
| MMLU | 82% | — |
Multimodal Not comparable
Amazon Nova Pro: 24.5 (#121), DeepSeek-R1-Distill-Llama-70B: —
| Benchmark | Amazon Nova Pro | DeepSeek-R1-Distill-Llama-70B |
|---|---|---|
| LMArena Vision | 980 | — |
Multilingual Not comparable
Amazon Nova Pro: 39.9 (#218), DeepSeek-R1-Distill-Llama-70B: —
| Benchmark | Amazon Nova Pro | DeepSeek-R1-Distill-Llama-70B |
|---|---|---|
| LMArena Non-English | 1234 | — |
| LMArena Chinese | 1244 | — |
| LMArena French | 1270 | — |
| LMArena German | 1243 | — |
| LMArena Japanese | 1200 | — |
| LMArena Korean | 1203 | — |
| LMArena Russian | 1240 | — |
| LMArena Spanish | 1182 | — |
Instruction Following DeepSeek-R1-Distill-Llama-70B leads
Amazon Nova Pro: 65.1 (#221), DeepSeek-R1-Distill-Llama-70B: 68.3 (#185)
| Benchmark | Amazon Nova Pro | DeepSeek-R1-Distill-Llama-70B |
|---|---|---|
| LiveBench Instruction Following | 67.1% | 69.9% |
| IFEval | 81.5% | — |
| LMArena Instruction Following | 1235 | — |
Long Context Not comparable
Amazon Nova Pro: 38.1 (#199), DeepSeek-R1-Distill-Llama-70B: —
| Benchmark | Amazon Nova Pro | DeepSeek-R1-Distill-Llama-70B |
|---|---|---|
| LMArena Longer Query | 1255 | — |
Writing & Preference DeepSeek-R1-Distill-Llama-70B leads
Amazon Nova Pro: 44.1 (#221), DeepSeek-R1-Distill-Llama-70B: 49.0 (#189)
| Benchmark | Amazon Nova Pro | DeepSeek-R1-Distill-Llama-70B |
|---|---|---|
| LiveBench Language | 37% | 23.8% |
| LMArena Text | 1259 | — |
| LMArena Creative Writing | 1212 | — |
| Short-Story Creative Writing | 60.5% | — |
| WildBench | 77.7% | — |
| LMArena Multi-Turn | 1245 | — |
Frequently asked questions
Is Amazon Nova Pro better than DeepSeek-R1-Distill-Llama-70B?
DeepSeek-R1-Distill-Llama-70B is the stronger model overall, scoring 37.4 to 30.5 on the Noometry Index.
Is Amazon Nova Pro or DeepSeek-R1-Distill-Llama-70B better for coding?
DeepSeek-R1-Distill-Llama-70B scores higher on coding benchmarks: 36.3 versus 34.6 in the Noometry coding category.
How many benchmarks do Amazon Nova Pro and DeepSeek-R1-Distill-Llama-70B share?
7 benchmarks have published results for both models. Amazon Nova Pro has 38 scored results on Noometry and DeepSeek-R1-Distill-Llama-70B has 13.