Model comparison
GPT-5.4 mini vs Granite 3.0 2b Instruct
GPT-5.4 mini is the stronger model overall, scoring 45.0 to 30.8 on the Noometry Index.
Last verified . 12 shared benchmarks.
Summary
- They share 12 benchmarks with published results for both. GPT-5.4 mini scores higher in 8 categories and Granite 3.0 2b Instruct in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GPT-5.4 mini leads 64.0 to 29.6.
- Granite 3.0 2b Instruct has downloadable open weights; the other is API-only.
Side by side
| GPT-5.4 mini | Granite 3.0 2b Instruct | |
|---|---|---|
| Provider | OpenAI | IBM |
| Noometry Index | 45.0 | 30.8 |
| Released | 2026-03-17 | — |
| Weights | Proprietary | Open |
| Context window | 400K | — |
| Max output | 128K | — |
| Input $ / M tokens | $0.75 | — |
| Output $ / M tokens | $4.50 | — |
| Results tracked | 46 | 13 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.4 mini leads
GPT-5.4 mini: 45.2 (#72), Granite 3.0 2b Instruct: 28.3 (#316)
| Benchmark | GPT-5.4 mini | Granite 3.0 2b Instruct |
|---|---|---|
| LMArena Coding | 1438 | 1090 |
| FrontierCode | 27% | — |
| LMArena WebDev | 1397 | — |
| SciCode | 49.9% | — |
| WeirdML | 60.3% | — |
| BigCodeBench Instruct | — | 20.5% |
| ALE-Bench | 1,189 | — |
Agentic & Tool Use Not comparable
GPT-5.4 mini: 29.9 (#81), Granite 3.0 2b Instruct: —
| Benchmark | GPT-5.4 mini | Granite 3.0 2b Instruct |
|---|---|---|
| DeepResearch Bench | 36.3% | — |
Reasoning GPT-5.4 mini leads
GPT-5.4 mini: 30.4 (#85), Granite 3.0 2b Instruct: 20.5 (#235)
| Benchmark | GPT-5.4 mini | Granite 3.0 2b Instruct |
|---|---|---|
| LMArena Hard Prompts | 1424 | 1073 |
| 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% | — |
| Mystery Game Puzzles | 11% | — |
| DTBench | 80% | — |
| LMCA | 40.8% | — |
| Epoch Capabilities Index | 148.84 | — |
| ForecastBench | 57 | — |
Math GPT-5.4 mini leads
GPT-5.4 mini: 45.5 (#75), Granite 3.0 2b Instruct: 32.2 (#217)
| Benchmark | GPT-5.4 mini | Granite 3.0 2b Instruct |
|---|---|---|
| LMArena Math | 1419 | 1117 |
| FrontierMath (Tiers 1-3) | 51.2% | — |
| FrontierMath Tier 4 | 9.8% | — |
| OTIS Mock AIME 2024-2025 | 88.9% | — |
| ProofBench | 21% | — |
| FrontierMath (Feb 2025 set) | 28.3% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge GPT-5.4 mini leads
GPT-5.4 mini: 51.5 (#67), Granite 3.0 2b Instruct: 29.0 (#241)
| Benchmark | GPT-5.4 mini | Granite 3.0 2b Instruct |
|---|---|---|
| LMArena Expert | 1435 | 1064 |
| GPQA Diamond | 86.9% | — |
| SimpleQA Verified | 29.4% | — |
| Vectara Hallucination Rate | 5.5% | — |
Multimodal Not comparable
GPT-5.4 mini: 39.7 (#56), Granite 3.0 2b Instruct: —
| Benchmark | GPT-5.4 mini | Granite 3.0 2b Instruct |
|---|---|---|
| LMArena Vision | 1245 | — |
Multilingual GPT-5.4 mini leads
GPT-5.4 mini: 51.9 (#96), Granite 3.0 2b Instruct: 27.0 (#278)
| Benchmark | GPT-5.4 mini | Granite 3.0 2b Instruct |
|---|---|---|
| LMArena Non-English | 1405 | 1033 |
| LMArena Chinese | 1446 | 1070 |
| LMArena Russian | 1417 | 1045 |
| LMArena French | 1440 | — |
| LMArena German | 1409 | — |
| LMArena Japanese | 1374 | — |
| LMArena Korean | 1368 | — |
| LMArena Spanish | 1405 | — |
Instruction Following GPT-5.4 mini leads
GPT-5.4 mini: 74.1 (#102), Granite 3.0 2b Instruct: 53.9 (#284)
| Benchmark | GPT-5.4 mini | Granite 3.0 2b Instruct |
|---|---|---|
| LMArena Instruction Following | 1405 | 1056 |
Long Context GPT-5.4 mini leads
GPT-5.4 mini: 43.0 (#112), Granite 3.0 2b Instruct: 32.5 (#268)
| Benchmark | GPT-5.4 mini | Granite 3.0 2b Instruct |
|---|---|---|
| LMArena Longer Query | 1407 | 1070 |
Writing & Preference GPT-5.4 mini leads
GPT-5.4 mini: 64.0 (#58), Granite 3.0 2b Instruct: 29.6 (#292)
| Benchmark | GPT-5.4 mini | Granite 3.0 2b Instruct |
|---|---|---|
| LMArena Text | 1412 | 1080 |
| LMArena Creative Writing | 1370 | 1046 |
| LMArena Multi-Turn | 1429 | 1053 |
| EQ-Bench Creative Writing | 1665 | — |
Frequently asked questions
Is GPT-5.4 mini better than Granite 3.0 2b Instruct?
GPT-5.4 mini is the stronger model overall, scoring 45.0 to 30.8 on the Noometry Index.
Is GPT-5.4 mini or Granite 3.0 2b Instruct better for coding?
GPT-5.4 mini scores higher on coding benchmarks: 45.2 versus 28.3 in the Noometry coding category.
How many benchmarks do GPT-5.4 mini and Granite 3.0 2b Instruct share?
12 benchmarks have published results for both models. GPT-5.4 mini has 46 scored results on Noometry and Granite 3.0 2b Instruct has 13.