Model comparison
GPT-5.4 nano vs Llama 2-70B
GPT-5.4 nano is the stronger model overall, scoring 41.9 to 24.1 on the Noometry Index.
Last verified . 22 shared benchmarks.
MetaSummary
- They share 22 benchmarks with published results for both. GPT-5.4 nano scores higher in 8 categories and Llama 2-70B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where GPT-5.4 nano leads 41.8 to 7.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 87.8% for GPT-5.4 nano and 0% for Llama 2-70B.
- Llama 2-70B has downloadable open weights; the other is API-only.
Side by side
| GPT-5.4 nano | Llama 2-70B | |
|---|---|---|
| Provider | OpenAI | Meta |
| Noometry Index | 41.9 | 24.1 |
| Released | 2026-03-17 | 2023-07-18 |
| Weights | Proprietary | Open |
| Context window | 400K | — |
| Max output | 128K | — |
| Input $ / M tokens | $0.20 | — |
| Output $ / M tokens | $1.25 | — |
| Results tracked | 40 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.4 nano leads
GPT-5.4 nano: 43.1 (#82), Llama 2-70B: 30.8 (#281)
| Benchmark | GPT-5.4 nano | Llama 2-70B |
|---|---|---|
| LMArena Coding | 1404 | 1079 |
| SciCode | 46.9% | — |
| WeirdML | 49.2% | — |
| ALE-Bench | 1,005 | — |
Reasoning GPT-5.4 nano leads
GPT-5.4 nano: 23.8 (#159), Llama 2-70B: 13.7 (#324)
| Benchmark | GPT-5.4 nano | Llama 2-70B |
|---|---|---|
| LMArena Hard Prompts | 1380 | 1073 |
| DTBench | 80.3% | 41.6% |
| Epoch Capabilities Index | 145.78 | 113.59 |
| ForecastBench | 57.3 | 51.4 |
| ARC-AGI-2 | 5.7% | — |
| Kagi LLM Benchmark | 39.7% | — |
| ARC-AGI-1 | 51.5% | — |
| CritPt | 9.3% | — |
| Chess Puzzles | 30% | — |
| Mystery Game Puzzles | 9% | — |
| LMCA | 36.9% | — |
| BIG-Bench Hard | — | 64.9% |
| CommonsenseQA 2.0 | — | 50% |
| HellaSwag | — | 85.3% |
| LAMBADA | — | 78.9% |
| PIQA | — | 82.8% |
| WinoGrande | — | 80.2% |
Math GPT-5.4 nano leads
GPT-5.4 nano: 40.7 (#76), Llama 2-70B: 7.5 (#320)
| Benchmark | GPT-5.4 nano | Llama 2-70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 87.8% | 0% |
| LMArena Math | 1406 | 1091 |
| FrontierMath (Tiers 1-3) | 44.9% | — |
| FrontierMath Tier 4 | 12.2% | — |
| ProofBench | 5% | — |
| MATH Level 5 | — | 3.3% |
| FrontierMath (Feb 2025 set) | 25.9% | — |
| FrontierMath Tier 4 (v1) | 6.3% | — |
| GSM8K | — | 69.6% |
Knowledge GPT-5.4 nano leads
GPT-5.4 nano: 41.8 (#94), Llama 2-70B: 7.4 (#305)
| Benchmark | GPT-5.4 nano | Llama 2-70B |
|---|---|---|
| GPQA Diamond | 78.5% | 26.3% |
| LMArena Expert | 1397 | 1039 |
| SimpleQA Verified | 11.7% | — |
| Vectara Hallucination Rate | 3.1% | — |
| ARC (AI2) Challenge | — | 78.3% |
| BoolQ | — | 88.6% |
| MMLU | — | 69.9% |
| OpenBookQA | — | 60.2% |
| TriviaQA | — | 87.6% |
Multimodal Not comparable
GPT-5.4 nano: 36.0 (#73), Llama 2-70B: —
| Benchmark | GPT-5.4 nano | Llama 2-70B |
|---|---|---|
| LMArena Vision | 1196 | — |
Multilingual GPT-5.4 nano leads
GPT-5.4 nano: 48.8 (#136), Llama 2-70B: 27.9 (#269)
| Benchmark | GPT-5.4 nano | Llama 2-70B |
|---|---|---|
| LMArena Non-English | 1359 | 1045 |
| LMArena Chinese | 1391 | 994 |
| LMArena French | 1396 | 1089 |
| LMArena German | 1368 | 1041 |
| LMArena Japanese | 1342 | 927 |
| LMArena Korean | 1321 | 964 |
| LMArena Russian | 1363 | 1083 |
| LMArena Spanish | 1368 | 1142 |
Instruction Following GPT-5.4 nano leads
GPT-5.4 nano: 72.1 (#139), Llama 2-70B: 55.1 (#273)
| Benchmark | GPT-5.4 nano | Llama 2-70B |
|---|---|---|
| LMArena Instruction Following | 1362 | 1071 |
Long Context GPT-5.4 nano leads
GPT-5.4 nano: 41.6 (#131), Llama 2-70B: 32.2 (#265)
| Benchmark | GPT-5.4 nano | Llama 2-70B |
|---|---|---|
| LMArena Longer Query | 1365 | 1062 |
Writing & Preference GPT-5.4 nano leads
GPT-5.4 nano: 55.9 (#137), Llama 2-70B: 32.5 (#274)
| Benchmark | GPT-5.4 nano | Llama 2-70B |
|---|---|---|
| LMArena Text | 1372 | 1115 |
| LMArena Creative Writing | 1313 | 1075 |
| LMArena Multi-Turn | 1382 | 1088 |
Frequently asked questions
Is GPT-5.4 nano better than Llama 2-70B?
GPT-5.4 nano is the stronger model overall, scoring 41.9 to 24.1 on the Noometry Index.
Is GPT-5.4 nano or Llama 2-70B better for coding?
GPT-5.4 nano scores higher on coding benchmarks: 43.1 versus 30.8 in the Noometry coding category.
How many benchmarks do GPT-5.4 nano and Llama 2-70B share?
22 benchmarks have published results for both models. GPT-5.4 nano has 40 scored results on Noometry and Llama 2-70B has 35.