Model comparison
GPT-5 Nano vs Granite 4.0 H Small
Granite 4.0 H Small is the stronger model overall, scoring 36.5 to 33.5 on the Noometry Index.
Last verified . 19 shared benchmarks.
Summary
- They share 19 benchmarks with published results for both. GPT-5 Nano scores higher in 3 categories and Granite 4.0 H Small in 5 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Granite 4.0 H Small leads 24.4 to 16.3.
- The biggest single-benchmark swing is GPQA (HELM): 67.9% for GPT-5 Nano and 38.3% for Granite 4.0 H Small.
- Granite 4.0 H Small has downloadable open weights; the other is API-only.
Side by side
| GPT-5 Nano | Granite 4.0 H Small | |
|---|---|---|
| Provider | OpenAI | IBM |
| Noometry Index | 33.5 | 36.5 |
| Released | 2025-08-07 | — |
| Weights | Proprietary | Open |
| Context window | 400K | — |
| Max output | 128K | — |
| Input $ / M tokens | $0.05 | — |
| Output $ / M tokens | $0.40 | — |
| Results tracked | 49 | 19 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Granite 4.0 H Small leads
GPT-5 Nano: 33.6 (#254), Granite 4.0 H Small: 36.4 (#209)
| Benchmark | GPT-5 Nano | Granite 4.0 H Small |
|---|---|---|
| LMArena Coding | 1351 | 1249 |
| SWE-bench Verified (bash only) | 34.8% | — |
| WeirdML | 38.1% | — |
| ALE-Bench | 718.67 | — |
Agentic & Tool Use Not comparable
GPT-5 Nano: 25.8 (#106), Granite 4.0 H Small: —
| Benchmark | GPT-5 Nano | Granite 4.0 H Small |
|---|---|---|
| Terminal-Bench | 21.8% | — |
| Berkeley Function Calling Leaderboard | 51.5% | — |
Reasoning Granite 4.0 H Small leads
GPT-5 Nano: 16.3 (#306), Granite 4.0 H Small: 24.4 (#163)
| Benchmark | GPT-5 Nano | Granite 4.0 H Small |
|---|---|---|
| LMArena Hard Prompts | 1328 | 1240 |
| ARC-AGI-2 | 2.6% | — |
| Kagi LLM Benchmark | 62.2% | — |
| ARC-AGI-1 | 20.7% | — |
| Chess Puzzles | 27% | — |
| Mystery Game Puzzles | 9% | — |
| DTBench | 62.7% | — |
| LMCA | 7.9% | — |
| Epoch Capabilities Index | 139.38 | — |
| ForecastBench | 59.1 | — |
Math Granite 4.0 H Small leads
GPT-5 Nano: 29.4 (#241), Granite 4.0 H Small: 31.4 (#223)
| Benchmark | GPT-5 Nano | Granite 4.0 H Small |
|---|---|---|
| Omni-MATH | 54.6% | 29.6% |
| LMArena Math | 1317 | 1247 |
| FrontierMath (Tiers 1-3) | 20% | — |
| FrontierMath Tier 4 | 2.4% | — |
| OTIS Mock AIME 2024-2025 | 81.1% | — |
| ProofBench | 12% | — |
| MATH Level 5 | 95.2% | — |
| FrontierMath (Feb 2025 set) | 8.3% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge GPT-5 Nano leads
GPT-5 Nano: 35.9 (#178), Granite 4.0 H Small: 32.0 (#215)
| Benchmark | GPT-5 Nano | Granite 4.0 H Small |
|---|---|---|
| MMLU-Pro | 77.8% | 56.9% |
| Vectara Hallucination Rate | 10.5% | 5.2% |
| GPQA (HELM) | 67.9% | 38.3% |
| LMArena Expert | 1321 | 1251 |
| GPQA Diamond | 69.4% | — |
| SimpleQA Verified | 11.7% | — |
Multimodal Not comparable
GPT-5 Nano: 31.3 (#108), Granite 4.0 H Small: —
| Benchmark | GPT-5 Nano | Granite 4.0 H Small |
|---|---|---|
| LMArena Vision | 1159 | — |
| VPCT | 37.2% | — |
Multilingual GPT-5 Nano leads
GPT-5 Nano: 45.3 (#172), Granite 4.0 H Small: 38.6 (#228)
| Benchmark | GPT-5 Nano | Granite 4.0 H Small |
|---|---|---|
| LMArena Non-English | 1313 | 1216 |
| LMArena Chinese | 1356 | 1249 |
| LMArena Russian | 1296 | 1201 |
| LMArena Spanish | 1360 | 1258 |
| LMArena German | 1327 | — |
| LMArena Japanese | 1226 | — |
| LMArena Korean | 1269 | — |
Instruction Following GPT-5 Nano leads
GPT-5 Nano: 75.0 (#79), Granite 4.0 H Small: 68.7 (#183)
| Benchmark | GPT-5 Nano | Granite 4.0 H Small |
|---|---|---|
| IFEval | 93.2% | 89% |
| LMArena Instruction Following | 1306 | 1222 |
Long Context Granite 4.0 H Small leads
GPT-5 Nano: 31.3 (#281), Granite 4.0 H Small: 37.7 (#213)
| Benchmark | GPT-5 Nano | Granite 4.0 H Small |
|---|---|---|
| LMArena Longer Query | 1312 | 1242 |
| Fiction.LiveBench | 44.4% | — |
Writing & Preference Granite 4.0 H Small leads
GPT-5 Nano: 39.1 (#249), Granite 4.0 H Small: 43.8 (#227)
| Benchmark | GPT-5 Nano | Granite 4.0 H Small |
|---|---|---|
| LMArena Text | 1320 | 1241 |
| LMArena Creative Writing | 1249 | 1211 |
| WildBench | 80.6% | 73.9% |
| LMArena Multi-Turn | 1311 | 1242 |
| EQ-Bench Creative Writing | 705 | — |
Frequently asked questions
Is GPT-5 Nano better than Granite 4.0 H Small?
Granite 4.0 H Small is the stronger model overall, scoring 36.5 to 33.5 on the Noometry Index.
Is GPT-5 Nano or Granite 4.0 H Small better for coding?
Granite 4.0 H Small scores higher on coding benchmarks: 36.4 versus 33.6 in the Noometry coding category.
How many benchmarks do GPT-5 Nano and Granite 4.0 H Small share?
19 benchmarks have published results for both models. GPT-5 Nano has 49 scored results on Noometry and Granite 4.0 H Small has 19.