Model comparison
GPT-5 Nano vs Llama-3.3-70B-Instruct
GPT-5 Nano is the stronger model overall, scoring 33.4 to 30.3 on the Noometry Index.
Last verified . 26 shared benchmarks.
MetaSummary
- They share 26 benchmarks with published results for both. GPT-5 Nano scores higher in 8 categories and Llama-3.3-70B-Instruct in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5 Nano leads 29.1 to 14.3.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 81.1% for GPT-5 Nano and 5.1% for Llama-3.3-70B-Instruct.
- GPT-5 Nano is cheaper at $0.05 / $0.40 per million input/output tokens, against $0.10 / $0.32 for Llama-3.3-70B-Instruct.
- GPT-5 Nano accepts more context: 400K tokens versus 128K.
- Llama-3.3-70B-Instruct has downloadable open weights; the other is API-only.
Side by side
| GPT-5 Nano | Llama-3.3-70B-Instruct | |
|---|---|---|
| Provider | OpenAI | Meta |
| Noometry Index | 33.4 | 30.3 |
| Released | 2025-08-07 | 2024-12-06 |
| Weights | Proprietary | Open |
| Context window | 400K | 128K |
| Max output | 128K | 4K |
| Input $ / M tokens | $0.05 | $0.10 |
| Output $ / M tokens | $0.40 | $0.32 |
| Results tracked | 48 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5 Nano leads
GPT-5 Nano: 33.3 (#248), Llama-3.3-70B-Instruct: 30.6 (#284)
| Benchmark | GPT-5 Nano | Llama-3.3-70B-Instruct |
|---|---|---|
| WeirdML | 38.1% | 14.4% |
| LMArena Coding | 1351 | 1268 |
| SWE-bench Verified (bash only) | 34.8% | — |
| SciCode | — | 26% |
| BigCodeBench Instruct | — | 46.9% |
| LiveBench Coding | — | 36.6% |
| BigCodeBench Complete | — | 57.5% |
| ALE-Bench | 718.67 | — |
Agentic & Tool Use Too close to call
GPT-5 Nano: 25.6 (#99), Llama-3.3-70B-Instruct: 25.6 (#100)
| Benchmark | GPT-5 Nano | Llama-3.3-70B-Instruct |
|---|---|---|
| Berkeley Function Calling Leaderboard | 51.5% | 31.9% |
| Terminal-Bench | 21.8% | — |
| BALROG | — | 23% |
Reasoning GPT-5 Nano leads
GPT-5 Nano: 16.4 (#291), Llama-3.3-70B-Instruct: 13.7 (#325)
| Benchmark | GPT-5 Nano | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Hard Prompts | 1329 | 1257 |
| DTBench | 62.7% | 59.5% |
| LMCA | 7.9% | 17.5% |
| Epoch Capabilities Index | 139.39 | 127.32 |
| ForecastBench | 59.1 | 58.6 |
| ARC-AGI-2 | 2.6% | — |
| SimpleBench | — | 19.9% |
| Kagi LLM Benchmark | 62.2% | — |
| ARC-AGI-1 | 20.7% | — |
| CritPt | — | 0% |
| Chess Puzzles | 27% | — |
| LiveBench Reasoning | — | 50.8% |
| Mystery Game Puzzles | 9% | — |
| LiveBench Data Analysis | — | 49.5% |
| LiveBench | — | 50.2% |
Math GPT-5 Nano leads
GPT-5 Nano: 29.1 (#221), Llama-3.3-70B-Instruct: 14.3 (#292)
| Benchmark | GPT-5 Nano | Llama-3.3-70B-Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 81.1% | 5.1% |
| LMArena Math | 1318 | 1268 |
| MATH Level 5 | 95.2% | 41.6% |
| FrontierMath (Tiers 1-3) | 20% | — |
| FrontierMath Tier 4 | 2.4% | — |
| ProofBench | 12% | — |
| Omni-MATH | 54.6% | — |
| LiveBench Math | — | 42.2% |
| FrontierMath (Feb 2025 set) | 8.3% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge GPT-5 Nano leads
GPT-5 Nano: 35.8 (#166), Llama-3.3-70B-Instruct: 30.5 (#220)
| Benchmark | GPT-5 Nano | Llama-3.3-70B-Instruct |
|---|---|---|
| GPQA Diamond | 69.4% | 47.4% |
| Vectara Hallucination Rate | 10.5% | 4.1% |
| LMArena Expert | 1321 | 1225 |
| SimpleQA Verified | 11.7% | — |
| MMLU-Pro | 77.8% | — |
| Confabulations | — | 22.8% |
| GPQA (HELM) | 67.9% | — |
| MMLU | — | 86.3% |
Multimodal Not comparable
GPT-5 Nano: 30.7 (#102), Llama-3.3-70B-Instruct: —
| Benchmark | GPT-5 Nano | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Vision | 1159 | — |
| VPCT | 37.2% | — |
Multilingual GPT-5 Nano leads
GPT-5 Nano: 45.5 (#167), Llama-3.3-70B-Instruct: 40.1 (#215)
| Benchmark | GPT-5 Nano | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Non-English | 1313 | 1236 |
| LMArena Chinese | 1355 | 1217 |
| LMArena German | 1329 | 1251 |
| LMArena Korean | 1269 | 1142 |
| LMArena Russian | 1296 | 1252 |
| LMArena Spanish | 1360 | 1269 |
| LMArena French | — | 1280 |
| LMArena Japanese | — | 1149 |
Instruction Following GPT-5 Nano leads
GPT-5 Nano: 75.2 (#72), Llama-3.3-70B-Instruct: 71.3 (#152)
| Benchmark | GPT-5 Nano | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Instruction Following | 1306 | 1242 |
| LiveBench Instruction Following | — | 82.7% |
| IFEval | 93.2% | — |
Long Context GPT-5 Nano leads
GPT-5 Nano: 31.3 (#275), Llama-3.3-70B-Instruct: 26.4 (#290)
| Benchmark | GPT-5 Nano | Llama-3.3-70B-Instruct |
|---|---|---|
| Fiction.LiveBench | 44.4% | 33.3% |
| LMArena Longer Query | 1313 | 1256 |
Writing & Preference Llama-3.3-70B-Instruct leads
GPT-5 Nano: 39.4 (#244), Llama-3.3-70B-Instruct: 47.8 (#203)
| Benchmark | GPT-5 Nano | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Text | 1320 | 1274 |
| LMArena Creative Writing | 1250 | 1250 |
| LMArena Multi-Turn | 1311 | 1279 |
| EQ-Bench Creative Writing | 705 | — |
| WildBench | 80.6% | — |
| LiveBench Language | — | 39.2% |
Frequently asked questions
Is GPT-5 Nano better than Llama-3.3-70B-Instruct?
GPT-5 Nano is the stronger model overall, scoring 33.4 to 30.3 on the Noometry Index.
Which is cheaper, GPT-5 Nano or Llama-3.3-70B-Instruct?
GPT-5 Nano is cheaper. It lists at $0.05 per million input tokens and $0.40 per million output tokens; Llama-3.3-70B-Instruct lists at $0.10 and $0.32.
Is GPT-5 Nano or Llama-3.3-70B-Instruct better for coding?
GPT-5 Nano scores higher on coding benchmarks: 33.3 versus 30.6 in the Noometry coding category.
Which has the bigger context window?
GPT-5 Nano does, with 400K tokens against 128K.
How many benchmarks do GPT-5 Nano and Llama-3.3-70B-Instruct share?
26 benchmarks have published results for both models. GPT-5 Nano has 48 scored results on Noometry and Llama-3.3-70B-Instruct has 43.