Model comparison
Llama 3.1-70B vs Llama 3.2 3B
Llama 3.1-70B is the stronger model overall, scoring 29.4 to 28.4 on the Noometry Index. Llama 3.2 3B costs 3.3× less per token, which makes it the better buy when Llama 3.1-70B's lead doesn't matter for your workload.
Last verified . 17 shared benchmarks.
MetaSummary
- They share 17 benchmarks with published results for both. Llama 3.1-70B scores higher in 7 categories and Llama 3.2 3B in 2 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Llama 3.2 3B leads 29.9 to 12.7.
- The biggest single-benchmark swing is BigCodeBench Complete: 54.8% for Llama 3.1-70B and 28.3% for Llama 3.2 3B.
- Llama 3.2 3B is cheaper at $0.05 / $0.33 per million input/output tokens, against $0.40 / $0.40 for Llama 3.1-70B.
- Llama 3.2 3B accepts more context: 131K tokens versus 128K.
Side by side
| Llama 3.1-70B | Llama 3.2 3B | |
|---|---|---|
| Provider | Meta | Meta |
| Noometry Index | 29.4 | 28.4 |
| Released | 2024-07-23 | 2024-09-24 |
| Weights | Open | Open |
| Context window | 128K | 131K |
| Max output | 4K | 118K |
| Input $ / M tokens | $0.40 | $0.05 |
| Output $ / M tokens | $0.40 | $0.33 |
| Results tracked | 35 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Llama 3.1-70B leads
Llama 3.1-70B: 29.8 (#292), Llama 3.2 3B: 27.1 (#314)
| Benchmark | Llama 3.1-70B | Llama 3.2 3B |
|---|---|---|
| BigCodeBench Instruct | 46.1% | 23.4% |
| LMArena Coding | 1260 | 1098 |
| BigCodeBench Complete | 54.8% | 28.3% |
| WeirdML | 9% | — |
Agentic & Tool Use Llama 3.1-70B leads
Llama 3.1-70B: 24.9 (#104), Llama 3.2 3B: 19.9 (#135)
| Benchmark | Llama 3.1-70B | Llama 3.2 3B |
|---|---|---|
| BALROG | 27.9% | 10.1% |
| Berkeley Function Calling Leaderboard | — | 21.9% |
| TheAgentCompany | 6.9% | — |
Reasoning Too close to call
Llama 3.1-70B: 20.8 (#215), Llama 3.2 3B: 20.1 (#225)
| Benchmark | Llama 3.1-70B | Llama 3.2 3B |
|---|---|---|
| LMArena Hard Prompts | 1241 | 1095 |
| DTBench | 60% | — |
| LMCA | 14.8% | — |
| Epoch Capabilities Index | 125.91 | — |
Math Llama 3.2 3B leads
Llama 3.1-70B: 12.7 (#298), Llama 3.2 3B: 29.9 (#213)
| Benchmark | Llama 3.1-70B | Llama 3.2 3B |
|---|---|---|
| LMArena Math | 1252 | 1126 |
| OTIS Mock AIME 2024-2025 | 3.6% | — |
| Omni-MATH | 21% | — |
| MATH Level 5 | 36.7% | — |
Knowledge Llama 3.2 3B leads
Llama 3.1-70B: 24.1 (#264), Llama 3.2 3B: 29.1 (#232)
| Benchmark | Llama 3.1-70B | Llama 3.2 3B |
|---|---|---|
| LMArena Expert | 1209 | 1090 |
| GPQA Diamond | 44.2% | — |
| MMLU-Pro | 65.3% | — |
| GPQA (HELM) | 42.6% | — |
| MMLU | 80.1% | — |
Multilingual Llama 3.1-70B leads
Llama 3.1-70B: 38.9 (#220), Llama 3.2 3B: 26.4 (#276)
| Benchmark | Llama 3.1-70B | Llama 3.2 3B |
|---|---|---|
| LMArena Non-English | 1219 | 1019 |
| LMArena Chinese | 1215 | 1017 |
| LMArena German | 1222 | 1056 |
| LMArena Russian | 1234 | 949 |
| LMArena French | 1260 | — |
| LMArena Japanese | 1132 | — |
| LMArena Korean | 1140 | — |
| LMArena Spanish | 1253 | — |
Instruction Following Llama 3.1-70B leads
Llama 3.1-70B: 65.5 (#218), Llama 3.2 3B: 56.3 (#270)
| Benchmark | Llama 3.1-70B | Llama 3.2 3B |
|---|---|---|
| LMArena Instruction Following | 1231 | 1089 |
| IFEval | 82.1% | — |
Long Context Llama 3.1-70B leads
Llama 3.1-70B: 37.6 (#208), Llama 3.2 3B: 33.4 (#256)
| Benchmark | Llama 3.1-70B | Llama 3.2 3B |
|---|---|---|
| LMArena Longer Query | 1241 | 1101 |
Writing & Preference Llama 3.1-70B leads
Llama 3.1-70B: 35.6 (#262), Llama 3.2 3B: 24.9 (#302)
| Benchmark | Llama 3.1-70B | Llama 3.2 3B |
|---|---|---|
| LMArena Text | 1261 | 1110 |
| LMArena Creative Writing | 1232 | 1094 |
| EQ-Bench Creative Writing | 784 | 595 |
| LMArena Multi-Turn | 1256 | 1105 |
| WildBench | 75.8% | — |
Frequently asked questions
Is Llama 3.1-70B better than Llama 3.2 3B?
Llama 3.1-70B is the stronger model overall, scoring 29.4 to 28.4 on the Noometry Index. Llama 3.2 3B costs 3.3× less per token, which makes it the better buy when Llama 3.1-70B's lead doesn't matter for your workload.
Which is cheaper, Llama 3.1-70B or Llama 3.2 3B?
Llama 3.2 3B is cheaper. It lists at $0.05 per million input tokens and $0.33 per million output tokens; Llama 3.1-70B lists at $0.40 and $0.40.
Is Llama 3.1-70B or Llama 3.2 3B better for coding?
Llama 3.1-70B scores higher on coding benchmarks: 29.8 versus 27.1 in the Noometry coding category.
Which has the bigger context window?
Llama 3.2 3B does, with 131K tokens against 128K.
How many benchmarks do Llama 3.1-70B and Llama 3.2 3B share?
17 benchmarks have published results for both models. Llama 3.1-70B has 35 scored results on Noometry and Llama 3.2 3B has 18.