Model comparison
Amazon Nova Micro vs GPT-5.4 mini
GPT-5.4 mini is the stronger model overall, scoring 44.8 to 29.9 on the Noometry Index. Amazon Nova Micro costs 28× less per token, which makes it the better buy when GPT-5.4 mini'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 Micro scores higher in 0 categories and GPT-5.4 mini in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GPT-5.4 mini leads 64.2 to 39.7.
- Amazon Nova Micro is cheaper at $0.035 / $0.14 per million input/output tokens, against $0.75 / $4.50 for GPT-5.4 mini.
- GPT-5.4 mini accepts more context: 400K tokens versus 128K.
Side by side
| Amazon Nova Micro | GPT-5.4 mini | |
|---|---|---|
| Provider | Amazon | OpenAI |
| Noometry Index | 29.9 | 44.8 |
| Released | 2024-12-03 | 2026-03-17 |
| Weights | Proprietary | Proprietary |
| Context window | 128K | 400K |
| Max output | 10K | 128K |
| Input $ / M tokens | $0.035 | $0.75 |
| Output $ / M tokens | $0.14 | $4.50 |
| Results tracked | 32 | 46 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.4 mini leads
Amazon Nova Micro: 30.0 (#289), GPT-5.4 mini: 45.0 (#67)
| Benchmark | Amazon Nova Micro | GPT-5.4 mini |
|---|---|---|
| LMArena Coding | 1218 | 1438 |
| FrontierCode | — | 27% |
| LMArena WebDev | — | 1397 |
| SciCode | — | 49.9% |
| WeirdML | — | 60.3% |
| LiveBench Coding | 20.2% | — |
| ALE-Bench | — | 1,189 |
Agentic & Tool Use GPT-5.4 mini leads
Amazon Nova Micro: 21.9 (#125), GPT-5.4 mini: 29.6 (#74)
| Benchmark | Amazon Nova Micro | GPT-5.4 mini |
|---|---|---|
| Berkeley Function Calling Leaderboard | 22.3% | — |
| DeepResearch Bench | — | 36.3% |
Reasoning GPT-5.4 mini leads
Amazon Nova Micro: 16.7 (#289), GPT-5.4 mini: 30.3 (#77)
| Benchmark | Amazon Nova Micro | GPT-5.4 mini |
|---|---|---|
| LMArena Hard Prompts | 1191 | 1424 |
| ARC-AGI-2 | — | 18.9% |
| Kagi LLM Benchmark | — | 37.9% |
| NYT Connections (extended) | — | 61.8% |
| ARC-AGI-1 | — | 63.7% |
| CritPt | — | 10% |
| Chess Puzzles | — | 24% |
| Thematic Generalization | — | 61.7% |
| LiveBench Reasoning | 25.1% | — |
| Mystery Game Puzzles | — | 11% |
| DTBench | — | 80% |
| LiveBench Data Analysis | 34% | — |
| LMCA | — | 40.8% |
| Epoch Capabilities Index | — | 148.83 |
| ForecastBench | — | 57 |
| LiveBench | 29.6% | — |
Math GPT-5.4 mini leads
Amazon Nova Micro: 24.9 (#252), GPT-5.4 mini: 45.4 (#63)
| Benchmark | Amazon Nova Micro | GPT-5.4 mini |
|---|---|---|
| LMArena Math | 1206 | 1418 |
| FrontierMath (Tiers 1-3) | — | 51.2% |
| FrontierMath Tier 4 | — | 9.8% |
| OTIS Mock AIME 2024-2025 | — | 88.9% |
| ProofBench | — | 21% |
| Omni-MATH | 21.4% | — |
| LiveBench Math | 34.5% | — |
| FrontierMath (Feb 2025 set) | — | 28.3% |
| FrontierMath Tier 4 (v1) | — | 2.1% |
Knowledge GPT-5.4 mini leads
Amazon Nova Micro: 29.2 (#231), GPT-5.4 mini: 51.4 (#62)
| Benchmark | Amazon Nova Micro | GPT-5.4 mini |
|---|---|---|
| Vectara Hallucination Rate | 5.5% | 5.5% |
| LMArena Expert | 1184 | 1434 |
| GPQA Diamond | — | 86.9% |
| SimpleQA Verified | — | 29.4% |
| MMLU-Pro | 51.1% | — |
| GPQA (HELM) | 38.3% | — |
| MMLU | 70.8% | — |
Multimodal Not comparable
Amazon Nova Micro: —, GPT-5.4 mini: 39.0 (#51)
| Benchmark | Amazon Nova Micro | GPT-5.4 mini |
|---|---|---|
| LMArena Vision | — | 1245 |
Multilingual GPT-5.4 mini leads
Amazon Nova Micro: 36.7 (#234), GPT-5.4 mini: 52.1 (#94)
| Benchmark | Amazon Nova Micro | GPT-5.4 mini |
|---|---|---|
| LMArena Non-English | 1186 | 1405 |
| LMArena Chinese | 1209 | 1446 |
| LMArena French | 1237 | 1440 |
| LMArena German | 1192 | 1407 |
| LMArena Japanese | 1154 | 1375 |
| LMArena Korean | 1150 | 1369 |
| LMArena Russian | 1185 | 1417 |
| LMArena Spanish | 1225 | 1403 |
Instruction Following GPT-5.4 mini leads
Amazon Nova Micro: 56.6 (#267), GPT-5.4 mini: 74.3 (#99)
| Benchmark | Amazon Nova Micro | GPT-5.4 mini |
|---|---|---|
| LMArena Instruction Following | 1174 | 1405 |
| LiveBench Instruction Following | 48% | — |
| IFEval | 76% | — |
Long Context GPT-5.4 mini leads
Amazon Nova Micro: 36.5 (#224), GPT-5.4 mini: 42.9 (#107)
| Benchmark | Amazon Nova Micro | GPT-5.4 mini |
|---|---|---|
| LMArena Longer Query | 1206 | 1407 |
Writing & Preference GPT-5.4 mini leads
Amazon Nova Micro: 39.7 (#242), GPT-5.4 mini: 64.2 (#56)
| Benchmark | Amazon Nova Micro | GPT-5.4 mini |
|---|---|---|
| LMArena Text | 1208 | 1411 |
| LMArena Creative Writing | 1172 | 1370 |
| LMArena Multi-Turn | 1178 | 1429 |
| EQ-Bench Creative Writing | — | 1665 |
| WildBench | 74.3% | — |
| LiveBench Language | 15.8% | — |
Frequently asked questions
Is Amazon Nova Micro better than GPT-5.4 mini?
GPT-5.4 mini is the stronger model overall, scoring 44.8 to 29.9 on the Noometry Index. Amazon Nova Micro costs 28× less per token, which makes it the better buy when GPT-5.4 mini's lead doesn't matter for your workload.
Which is cheaper, Amazon Nova Micro or GPT-5.4 mini?
Amazon Nova Micro is cheaper. It lists at $0.035 per million input tokens and $0.14 per million output tokens; GPT-5.4 mini lists at $0.75 and $4.50.
Is Amazon Nova Micro or GPT-5.4 mini better for coding?
GPT-5.4 mini scores higher on coding benchmarks: 45.0 versus 30.0 in the Noometry coding category.
Which has the bigger context window?
GPT-5.4 mini does, with 400K tokens against 128K.
How many benchmarks do Amazon Nova Micro and GPT-5.4 mini share?
18 benchmarks have published results for both models. Amazon Nova Micro has 32 scored results on Noometry and GPT-5.4 mini has 46.