Model comparison
Amazon Nova Pro vs Kimi K3
Kimi K3 is the stronger model overall, scoring 59.5 to 31.0 on the Noometry Index. Amazon Nova Pro costs 4.3× less per token, which makes it the better buy when Kimi K3's lead doesn't matter for your workload.
Last verified . 18 shared benchmarks.
Summary
- They share 18 benchmarks with published results for both. Amazon Nova Pro scores higher in 0 categories and Kimi K3 in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K3 leads 74.2 to 28.5.
- Amazon Nova Pro is cheaper at $0.80 / $3.20 per million input/output tokens, against $3 / $15 for Kimi K3.
- Kimi K3 accepts more context: 1.05M tokens versus 300K.
- Kimi K3 has downloadable open weights; the other is API-only.
Side by side
| Amazon Nova Pro | Kimi K3 | |
|---|---|---|
| Provider | Amazon | Moonshot AI |
| Noometry Index | 31.0 | 59.5 |
| Released | 2024-12-03 | 2026-07-16 |
| Weights | Proprietary | Open |
| Context window | 300K | 1.05M |
| Max output | 10K | 1.05M |
| Input $ / M tokens | $0.80 | $3 |
| Output $ / M tokens | $3.20 | $15 |
| Results tracked | 38 | 53 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K3 leads
Amazon Nova Pro: 35.1 (#229), Kimi K3: 61.0 (#10)
| Benchmark | Amazon Nova Pro | Kimi K3 |
|---|---|---|
| LMArena Coding | 1270 | 1508 |
| DeepSWE | — | 68.5% |
| FrontierCode | — | 44.2% |
| LMArena WebDev | — | 1654 |
| FrontierSWE | — | 25.9% |
| SciCode | — | 59.5% |
| WeirdML | — | 82.6% |
| LiveBench Coding | 38.1% | — |
| ALE-Bench | — | 1,524 |
Agentic & Tool Use Kimi K3 leads
Amazon Nova Pro: 16.7 (#147), Kimi K3: 41.8 (#20)
| Benchmark | Amazon Nova Pro | Kimi K3 |
|---|---|---|
| APEX-Agents | — | 50.6% |
| Berkeley Function Calling Leaderboard | 25% | — |
| TheAgentCompany | 1.7% | — |
| τ²-bench Banking | — | 37.1% |
| PostTrainBench | — | 32% |
| GBAEval | — | 48.3% |
| GDP.pdf | — | 19% |
| Vending-Bench 2 | — | 5,165 |
Reasoning Kimi K3 leads
Amazon Nova Pro: 20.0 (#243), Kimi K3: 63.0 (#17)
| Benchmark | Amazon Nova Pro | Kimi K3 |
|---|---|---|
| LMArena Hard Prompts | 1246 | 1496 |
| Epoch Capabilities Index | 123.8 | 157.45 |
| ARC-AGI-2 | — | 60.4% |
| SimpleBench | — | 60.7% |
| NYT Connections (extended) | — | 93.6% |
| ARC-AGI-1 | — | 94.5% |
| CritPt | — | 23.4% |
| Chess Puzzles | — | 39% |
| LiveBench Reasoning | 32.6% | — |
| Mystery Game Puzzles | — | 26% |
| DTBench | — | 91.2% |
| LiveBench Data Analysis | 48.3% | — |
| LMCA | — | 52.7% |
| Surface Evolver Bench | — | 95% |
| ForecastBench | — | 61.1 |
| LiveBench | 43.5% | — |
Math Kimi K3 leads
Amazon Nova Pro: 28.5 (#243), Kimi K3: 74.2 (#16)
| Benchmark | Amazon Nova Pro | Kimi K3 |
|---|---|---|
| LMArena Math | 1252 | 1491 |
| FrontierMath (Tiers 1-3) | — | 72.2% |
| FrontierMath Tier 4 | — | 39% |
| MathArena Final-Answer Competitions | — | 87.8% |
| OTIS Mock AIME 2024-2025 | — | 97.2% |
| ProofBench | — | 87% |
| Omni-MATH | 24.2% | — |
| LiveBench Math | 38% | — |
Knowledge Kimi K3 leads
Amazon Nova Pro: 27.4 (#250), Kimi K3: 63.2 (#21)
| Benchmark | Amazon Nova Pro | Kimi K3 |
|---|---|---|
| LMArena Expert | 1211 | 1521 |
| GPQA Diamond | — | 93.1% |
| Humanity's Last Exam | 4.4% | — |
| SimpleQA Verified | — | 50.6% |
| MMLU-Pro | 67.3% | — |
| Confabulations | 30.1% | — |
| Vectara Hallucination Rate | 5.1% | — |
| GPQA (HELM) | 44.6% | — |
| MMLU | 82% | — |
Multimodal Kimi K3 leads
Amazon Nova Pro: 25.0 (#126), Kimi K3: 37.8 (#70)
| Benchmark | Amazon Nova Pro | Kimi K3 |
|---|---|---|
| LMArena Vision | 980 | — |
| Blueprint-Bench 2 | — | 29.5% |
| Furniture Assembly | — | 34.2% |
Multilingual Kimi K3 leads
Amazon Nova Pro: 39.7 (#223), Kimi K3: 56.3 (#21)
| Benchmark | Amazon Nova Pro | Kimi K3 |
|---|---|---|
| LMArena Non-English | 1234 | 1466 |
| LMArena Chinese | 1244 | 1529 |
| LMArena French | 1271 | 1491 |
| LMArena German | 1243 | 1488 |
| LMArena Japanese | 1200 | 1487 |
| LMArena Korean | 1203 | 1458 |
| LMArena Russian | 1240 | 1482 |
| LMArena Spanish | 1182 | 1472 |
Instruction Following Kimi K3 leads
Amazon Nova Pro: 64.9 (#226), Kimi K3: 77.7 (#14)
| Benchmark | Amazon Nova Pro | Kimi K3 |
|---|---|---|
| LMArena Instruction Following | 1235 | 1483 |
| LiveBench Instruction Following | 67.1% | — |
| IFEval | 81.5% | — |
Long Context Kimi K3 leads
Amazon Nova Pro: 38.1 (#205), Kimi K3: 45.8 (#29)
| Benchmark | Amazon Nova Pro | Kimi K3 |
|---|---|---|
| LMArena Longer Query | 1255 | 1494 |
Writing & Preference Kimi K3 leads
Amazon Nova Pro: 43.9 (#226), Kimi K3: 76.6 (#4)
| Benchmark | Amazon Nova Pro | Kimi K3 |
|---|---|---|
| LMArena Text | 1259 | 1476 |
| LMArena Creative Writing | 1212 | 1454 |
| LMArena Multi-Turn | 1246 | 1488 |
| Short-Story Creative Writing | 60.5% | — |
| EQ-Bench Creative Writing | — | 2082 |
| WildBench | 77.7% | — |
| EQ-Bench 4 | — | 1339 |
| LiveBench Language | 37% | — |
Frequently asked questions
Is Amazon Nova Pro better than Kimi K3?
Kimi K3 is the stronger model overall, scoring 59.5 to 31.0 on the Noometry Index. Amazon Nova Pro costs 4.3× less per token, which makes it the better buy when Kimi K3's lead doesn't matter for your workload.
Which is cheaper, Amazon Nova Pro or Kimi K3?
Amazon Nova Pro is cheaper. It lists at $0.80 per million input tokens and $3.20 per million output tokens; Kimi K3 lists at $3 and $15.
Is Amazon Nova Pro or Kimi K3 better for coding?
Kimi K3 scores higher on coding benchmarks: 61.0 versus 35.1 in the Noometry coding category.
Which has the bigger context window?
Kimi K3 does, with 1.05M tokens against 300K.
How many benchmarks do Amazon Nova Pro and Kimi K3 share?
18 benchmarks have published results for both models. Amazon Nova Pro has 38 scored results on Noometry and Kimi K3 has 53.