Model comparison
GPT-5.4 nano vs Llama-3.3-70B-Instruct
GPT-5.4 nano is the stronger model overall, scoring 41.9 to 30.6 on the Noometry Index. Llama-3.3-70B-Instruct costs 3.0× less per token, which makes it the better buy when GPT-5.4 nano's lead doesn't matter for your workload.
Last verified . 27 shared benchmarks.
MetaSummary
- They share 27 benchmarks with published results for both. GPT-5.4 nano scores higher in 8 categories and Llama-3.3-70B-Instruct in 0 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.4 nano leads 40.9 to 15.3.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 87.8% for GPT-5.4 nano and 5.1% for Llama-3.3-70B-Instruct.
- Llama-3.3-70B-Instruct is cheaper at $0.10 / $0.32 per million input/output tokens, against $0.20 / $1.25 for GPT-5.4 nano.
- GPT-5.4 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-5.4 nano | Llama-3.3-70B-Instruct | |
|---|---|---|
| Provider | OpenAI | Meta |
| Noometry Index | 41.9 | 30.6 |
| Released | 2026-03-17 | 2024-12-06 |
| Weights | Proprietary | Open |
| Context window | 400K | 128K |
| Max output | 128K | 4K |
| Input $ / M tokens | $0.20 | $0.10 |
| Output $ / M tokens | $1.25 | $0.32 |
| Results tracked | 40 | 43 |
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: 43.6 (#84), Llama-3.3-70B-Instruct: 31.0 (#290)
| Benchmark | GPT-5.4 nano | Llama-3.3-70B-Instruct |
|---|---|---|
| SciCode | 46.9% | 26% |
| WeirdML | 49.2% | 14.4% |
| LMArena Coding | 1405 | 1268 |
| BigCodeBench Instruct | — | 46.9% |
| LiveBench Coding | — | 36.6% |
| BigCodeBench Complete | — | 57.5% |
| ALE-Bench | 1,005 | — |
Agentic & Tool Use Not comparable
GPT-5.4 nano: —, Llama-3.3-70B-Instruct: 25.8 (#105)
| Benchmark | GPT-5.4 nano | Llama-3.3-70B-Instruct |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 31.9% |
| BALROG | — | 23% |
Reasoning GPT-5.4 nano leads
GPT-5.4 nano: 23.7 (#173), Llama-3.3-70B-Instruct: 14.1 (#327)
| Benchmark | GPT-5.4 nano | Llama-3.3-70B-Instruct |
|---|---|---|
| CritPt | 9.3% | 0% |
| LMArena Hard Prompts | 1381 | 1257 |
| DTBench | 80.3% | 59.5% |
| LMCA | 36.9% | 17.5% |
| Epoch Capabilities Index | 145.81 | 127.33 |
| ForecastBench | 57.3 | 58.6 |
| ARC-AGI-2 | 5.7% | — |
| SimpleBench | — | 19.9% |
| Kagi LLM Benchmark | 39.7% | — |
| ARC-AGI-1 | 51.5% | — |
| Chess Puzzles | 30% | — |
| LiveBench Reasoning | — | 50.8% |
| Mystery Game Puzzles | 9% | — |
| LiveBench Data Analysis | — | 49.5% |
| LiveBench | — | 50.2% |
Math GPT-5.4 nano leads
GPT-5.4 nano: 40.9 (#88), Llama-3.3-70B-Instruct: 15.3 (#298)
| Benchmark | GPT-5.4 nano | Llama-3.3-70B-Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 87.8% | 5.1% |
| LMArena Math | 1406 | 1267 |
| FrontierMath (Tiers 1-3) | 44.9% | — |
| FrontierMath Tier 4 | 12.2% | — |
| ProofBench | 5% | — |
| LiveBench Math | — | 42.2% |
| MATH Level 5 | — | 41.6% |
| FrontierMath (Feb 2025 set) | 25.9% | — |
| FrontierMath Tier 4 (v1) | 6.3% | — |
Knowledge GPT-5.4 nano leads
GPT-5.4 nano: 41.9 (#103), Llama-3.3-70B-Instruct: 30.6 (#226)
| Benchmark | GPT-5.4 nano | Llama-3.3-70B-Instruct |
|---|---|---|
| GPQA Diamond | 78.5% | 47.4% |
| Vectara Hallucination Rate | 3.1% | 4.1% |
| LMArena Expert | 1396 | 1225 |
| SimpleQA Verified | 11.7% | — |
| Confabulations | — | 22.8% |
| MMLU | — | 86.3% |
Multimodal Not comparable
GPT-5.4 nano: 36.7 (#78), Llama-3.3-70B-Instruct: —
| Benchmark | GPT-5.4 nano | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Vision | 1196 | — |
Multilingual GPT-5.4 nano leads
GPT-5.4 nano: 48.6 (#140), Llama-3.3-70B-Instruct: 39.9 (#220)
| Benchmark | GPT-5.4 nano | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Non-English | 1359 | 1236 |
| LMArena Chinese | 1392 | 1217 |
| LMArena French | 1396 | 1281 |
| LMArena German | 1367 | 1251 |
| LMArena Japanese | 1343 | 1150 |
| LMArena Korean | 1320 | 1143 |
| LMArena Russian | 1363 | 1252 |
| LMArena Spanish | 1371 | 1270 |
Instruction Following Too close to call
GPT-5.4 nano: 71.9 (#144), Llama-3.3-70B-Instruct: 71.1 (#157)
| Benchmark | GPT-5.4 nano | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Instruction Following | 1362 | 1242 |
| LiveBench Instruction Following | — | 82.7% |
Long Context GPT-5.4 nano leads
GPT-5.4 nano: 41.6 (#137), Llama-3.3-70B-Instruct: 26.4 (#295)
| Benchmark | GPT-5.4 nano | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Longer Query | 1366 | 1256 |
| Fiction.LiveBench | — | 33.3% |
Writing & Preference GPT-5.4 nano leads
GPT-5.4 nano: 55.7 (#142), Llama-3.3-70B-Instruct: 47.6 (#207)
| Benchmark | GPT-5.4 nano | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Text | 1372 | 1274 |
| LMArena Creative Writing | 1314 | 1250 |
| LMArena Multi-Turn | 1382 | 1280 |
| LiveBench Language | — | 39.2% |
Frequently asked questions
Is GPT-5.4 nano better than Llama-3.3-70B-Instruct?
GPT-5.4 nano is the stronger model overall, scoring 41.9 to 30.6 on the Noometry Index. Llama-3.3-70B-Instruct costs 3.0× less per token, which makes it the better buy when GPT-5.4 nano's lead doesn't matter for your workload.
Which is cheaper, GPT-5.4 nano or Llama-3.3-70B-Instruct?
Llama-3.3-70B-Instruct is cheaper. It lists at $0.10 per million input tokens and $0.32 per million output tokens; GPT-5.4 nano lists at $0.20 and $1.25.
Is GPT-5.4 nano or Llama-3.3-70B-Instruct better for coding?
GPT-5.4 nano scores higher on coding benchmarks: 43.6 versus 31.0 in the Noometry coding category.
Which has the bigger context window?
GPT-5.4 nano does, with 400K tokens against 128K.
How many benchmarks do GPT-5.4 nano and Llama-3.3-70B-Instruct share?
27 benchmarks have published results for both models. GPT-5.4 nano has 40 scored results on Noometry and Llama-3.3-70B-Instruct has 43.