Model comparison
GPT-5.4 Pro vs Llama 3.2 1B
GPT-5.4 Pro is the stronger model overall, scoring 58.9 to 20.1 on the Noometry Index. Llama 3.2 1B costs 957× less per token, which makes it the better buy when GPT-5.4 Pro's lead doesn't matter for your workload.
Last verified . 3 shared benchmarks.
MetaSummary
- They share 3 benchmarks with published results for both. GPT-5.4 Pro scores higher in 4 categories and Llama 3.2 1B in 0 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.4 Pro leads 72.4 to 10.4.
- The biggest single-benchmark swing is GPQA Diamond: 94.6% for GPT-5.4 Pro and 23.9% for Llama 3.2 1B.
- Llama 3.2 1B is cheaper at $0.027 / $0.20 per million input/output tokens, against $30 / $180 for GPT-5.4 Pro.
- GPT-5.4 Pro accepts more context: 1.05M tokens versus 60K.
- Llama 3.2 1B has downloadable open weights; the other is API-only.
Side by side
| GPT-5.4 Pro | Llama 3.2 1B | |
|---|---|---|
| Provider | OpenAI | Meta |
| Noometry Index | 58.9 | 20.1 |
| Released | 2026-03-05 | 2024-09-24 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 60K |
| Max output | 128K | 54K |
| Input $ / M tokens | $30 | $0.027 |
| Output $ / M tokens | $180 | $0.20 |
| Results tracked | 16 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.4 Pro leads
GPT-5.4 Pro: 43.2 (#90), Llama 3.2 1B: 21.1 (#338)
| Benchmark | GPT-5.4 Pro | Llama 3.2 1B |
|---|---|---|
| WeirdML | 57.4% | — |
| BigCodeBench Instruct | — | 8.2% |
| LMArena Coding | — | 1070 |
| BigCodeBench Complete | — | 11.3% |
Agentic & Tool Use Not comparable
GPT-5.4 Pro: —, Llama 3.2 1B: 14.6 (#150)
| Benchmark | GPT-5.4 Pro | Llama 3.2 1B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 10.8% |
| BALROG | — | 6.6% |
Reasoning GPT-5.4 Pro leads
GPT-5.4 Pro: 70.7 (#13), Llama 3.2 1B: 16.2 (#308)
| Benchmark | GPT-5.4 Pro | Llama 3.2 1B |
|---|---|---|
| Chess Puzzles | 58.6% | 0% |
| Epoch Capabilities Index | 158.93 | 101.99 |
| ARC-AGI-2 | 83.3% | — |
| SimpleBench | 74.1% | — |
| ARC-AGI-1 | 94.5% | — |
| CritPt | 30% | — |
| EnigmaEval | 23.8% | — |
| LMArena Hard Prompts | — | 1044 |
Math GPT-5.4 Pro leads
GPT-5.4 Pro: 72.4 (#22), Llama 3.2 1B: 10.4 (#313)
| Benchmark | GPT-5.4 Pro | Llama 3.2 1B |
|---|---|---|
| FrontierMath (Tiers 1-3) | 82.5% | — |
| FrontierMath Tier 4 | 58.5% | — |
| OTIS Mock AIME 2024-2025 | — | 0.6% |
| LMArena Math | — | 1086 |
| FrontierMath (Feb 2025 set) | 50% | — |
| FrontierMath Tier 4 (v1) | 37.5% | — |
Knowledge GPT-5.4 Pro leads
GPT-5.4 Pro: 68.3 (#7), Llama 3.2 1B: 7.2 (#312)
| Benchmark | GPT-5.4 Pro | Llama 3.2 1B |
|---|---|---|
| GPQA Diamond | 94.6% | 23.9% |
| Humanity's Last Exam | 44.3% | — |
| SimpleQA Verified | 46.3% | — |
| Vectara Hallucination Rate | 8.3% | — |
| LMArena Expert | — | 1007 |
Multilingual Not comparable
GPT-5.4 Pro: —, Llama 3.2 1B: 23.8 (#292)
| Benchmark | GPT-5.4 Pro | Llama 3.2 1B |
|---|---|---|
| LMArena Non-English | — | 973 |
| LMArena Chinese | — | 959 |
| LMArena German | — | 1014 |
| LMArena Russian | — | 941 |
Instruction Following Not comparable
GPT-5.4 Pro: —, Llama 3.2 1B: 52.4 (#290)
| Benchmark | GPT-5.4 Pro | Llama 3.2 1B |
|---|---|---|
| LMArena Instruction Following | — | 1031 |
Long Context Not comparable
GPT-5.4 Pro: —, Llama 3.2 1B: 31.9 (#274)
| Benchmark | GPT-5.4 Pro | Llama 3.2 1B |
|---|---|---|
| LMArena Longer Query | — | 1050 |
Writing & Preference Not comparable
GPT-5.4 Pro: —, Llama 3.2 1B: 21.3 (#310)
| Benchmark | GPT-5.4 Pro | Llama 3.2 1B |
|---|---|---|
| LMArena Text | — | 1055 |
| LMArena Creative Writing | — | 1033 |
| EQ-Bench Creative Writing | — | 200 |
| LMArena Multi-Turn | — | 1030 |
Frequently asked questions
Is GPT-5.4 Pro better than Llama 3.2 1B?
GPT-5.4 Pro is the stronger model overall, scoring 58.9 to 20.1 on the Noometry Index. Llama 3.2 1B costs 957× less per token, which makes it the better buy when GPT-5.4 Pro's lead doesn't matter for your workload.
Which is cheaper, GPT-5.4 Pro or Llama 3.2 1B?
Llama 3.2 1B is cheaper. It lists at $0.027 per million input tokens and $0.20 per million output tokens; GPT-5.4 Pro lists at $30 and $180.
Is GPT-5.4 Pro or Llama 3.2 1B better for coding?
GPT-5.4 Pro scores higher on coding benchmarks: 43.2 versus 21.1 in the Noometry coding category.
Which has the bigger context window?
GPT-5.4 Pro does, with 1.05M tokens against 60K.
How many benchmarks do GPT-5.4 Pro and Llama 3.2 1B share?
3 benchmarks have published results for both models. GPT-5.4 Pro has 16 scored results on Noometry and Llama 3.2 1B has 22.