Model comparison
Gemma 2 27B vs Llama-3.3-70B-Instruct
Llama-3.3-70B-Instruct is the stronger model overall, scoring 30.6 to 29.4 on the Noometry Index.
Last verified . 34 shared benchmarks.
Google
MetaSummary
- They share 34 benchmarks with published results for both. Gemma 2 27B scores higher in 3 categories and Llama-3.3-70B-Instruct in 5 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Llama-3.3-70B-Instruct leads 30.6 to 19.0.
- The biggest single-benchmark swing is LiveBench Instruction Following: 58.1% for Gemma 2 27B and 82.7% for Llama-3.3-70B-Instruct.
- Llama-3.3-70B-Instruct is cheaper at $0.10 / $0.32 per million input/output tokens, against $0.65 / $0.65 for Gemma 2 27B.
- Llama-3.3-70B-Instruct accepts more context: 128K tokens versus 8K.
Side by side
| Gemma 2 27B | Llama-3.3-70B-Instruct | |
|---|---|---|
| Provider | Meta | |
| Noometry Index | 29.4 | 30.6 |
| Released | 2024-06-24 | 2024-12-06 |
| Weights | Open | Open |
| Context window | 8K | 128K |
| Max output | 2K | 4K |
| Input $ / M tokens | $0.65 | $0.10 |
| Output $ / M tokens | $0.65 | $0.32 |
| Results tracked | 34 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 2 27B leads
Gemma 2 27B: 34.1 (#246), Llama-3.3-70B-Instruct: 31.0 (#290)
| Benchmark | Gemma 2 27B | Llama-3.3-70B-Instruct |
|---|---|---|
| BigCodeBench Instruct | 42.8% | 46.9% |
| LiveBench Coding | 36% | 36.6% |
| LMArena Coding | 1211 | 1268 |
| BigCodeBench Complete | 52.5% | 57.5% |
| SciCode | — | 26% |
| WeirdML | — | 14.4% |
Agentic & Tool Use Not comparable
Gemma 2 27B: —, Llama-3.3-70B-Instruct: 25.8 (#105)
| Benchmark | Gemma 2 27B | Llama-3.3-70B-Instruct |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 31.9% |
| BALROG | — | 23% |
Reasoning Gemma 2 27B leads
Gemma 2 27B: 15.3 (#315), Llama-3.3-70B-Instruct: 14.1 (#327)
| Benchmark | Gemma 2 27B | Llama-3.3-70B-Instruct |
|---|---|---|
| LiveBench Reasoning | 28.1% | 50.8% |
| LMArena Hard Prompts | 1198 | 1257 |
| DTBench | 48% | 59.5% |
| LiveBench Data Analysis | 47.9% | 49.5% |
| LMCA | 7.1% | 17.5% |
| Epoch Capabilities Index | 122.08 | 127.33 |
| LiveBench | 38.2% | 50.2% |
| SimpleBench | — | 19.9% |
| CritPt | — | 0% |
| ForecastBench | — | 58.6 |
Math Llama-3.3-70B-Instruct leads
Gemma 2 27B: 10.7 (#311), Llama-3.3-70B-Instruct: 15.3 (#298)
| Benchmark | Gemma 2 27B | Llama-3.3-70B-Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.4% | 5.1% |
| LiveBench Math | 26.5% | 42.2% |
| LMArena Math | 1212 | 1267 |
| MATH Level 5 | 27.9% | 41.6% |
Knowledge Llama-3.3-70B-Instruct leads
Gemma 2 27B: 19.0 (#280), Llama-3.3-70B-Instruct: 30.6 (#226)
| Benchmark | Gemma 2 27B | Llama-3.3-70B-Instruct |
|---|---|---|
| GPQA Diamond | 36.5% | 47.4% |
| Confabulations | 27.1% | 22.8% |
| LMArena Expert | 1172 | 1225 |
| MMLU | 75.7% | 86.3% |
| Vectara Hallucination Rate | — | 4.1% |
Multilingual Llama-3.3-70B-Instruct leads
Gemma 2 27B: 38.6 (#226), Llama-3.3-70B-Instruct: 39.9 (#220)
| Benchmark | Gemma 2 27B | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Non-English | 1217 | 1236 |
| LMArena Chinese | 1221 | 1217 |
| LMArena French | 1247 | 1281 |
| LMArena German | 1209 | 1251 |
| LMArena Japanese | 1175 | 1150 |
| LMArena Korean | 1174 | 1143 |
| LMArena Russian | 1234 | 1252 |
| LMArena Spanish | 1228 | 1270 |
Instruction Following Llama-3.3-70B-Instruct leads
Gemma 2 27B: 60.5 (#249), Llama-3.3-70B-Instruct: 71.1 (#157)
| Benchmark | Gemma 2 27B | Llama-3.3-70B-Instruct |
|---|---|---|
| LiveBench Instruction Following | 58.1% | 82.7% |
| LMArena Instruction Following | 1206 | 1242 |
Long Context Gemma 2 27B leads
Gemma 2 27B: 37.3 (#218), Llama-3.3-70B-Instruct: 26.4 (#295)
| Benchmark | Gemma 2 27B | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Longer Query | 1231 | 1256 |
| Fiction.LiveBench | — | 33.3% |
Writing & Preference Llama-3.3-70B-Instruct leads
Gemma 2 27B: 44.2 (#225), Llama-3.3-70B-Instruct: 47.6 (#207)
| Benchmark | Gemma 2 27B | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Text | 1231 | 1274 |
| LMArena Creative Writing | 1241 | 1250 |
| LMArena Multi-Turn | 1224 | 1280 |
| LiveBench Language | 32.6% | 39.2% |
Frequently asked questions
Is Gemma 2 27B better than Llama-3.3-70B-Instruct?
Llama-3.3-70B-Instruct is the stronger model overall, scoring 30.6 to 29.4 on the Noometry Index.
Which is cheaper, Gemma 2 27B 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; Gemma 2 27B lists at $0.65 and $0.65.
Is Gemma 2 27B or Llama-3.3-70B-Instruct better for coding?
Gemma 2 27B scores higher on coding benchmarks: 34.1 versus 31.0 in the Noometry coding category.
Which has the bigger context window?
Llama-3.3-70B-Instruct does, with 128K tokens against 8K.
How many benchmarks do Gemma 2 27B and Llama-3.3-70B-Instruct share?
34 benchmarks have published results for both models. Gemma 2 27B has 34 scored results on Noometry and Llama-3.3-70B-Instruct has 43.