Model comparison
Gemma 4 31B IT vs Llama 3.1-70B
Gemma 4 31B IT is the stronger model overall, scoring 43.2 to 29.4 on the Noometry Index.
Last verified . 21 shared benchmarks.
Google
MetaSummary
- They share 21 benchmarks with published results for both. Gemma 4 31B IT scores higher in 8 categories and Llama 3.1-70B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Gemma 4 31B IT leads 41.3 to 12.7.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 73.3% for Gemma 4 31B IT and 3.6% for Llama 3.1-70B.
- Gemma 4 31B IT is cheaper at $0.09 / $0.34 per million input/output tokens, against $0.40 / $0.40 for Llama 3.1-70B.
- Gemma 4 31B IT accepts more context: 262K tokens versus 128K.
Side by side
| Gemma 4 31B IT | Llama 3.1-70B | |
|---|---|---|
| Provider | Meta | |
| Noometry Index | 43.2 | 29.4 |
| Released | 2026-04-02 | 2024-07-23 |
| Weights | Open | Open |
| Context window | 262K | 128K |
| Max output | 33K | 4K |
| Input $ / M tokens | $0.09 | $0.40 |
| Output $ / M tokens | $0.34 | $0.40 |
| Results tracked | 35 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 4 31B IT leads
Gemma 4 31B IT: 41.9 (#98), Llama 3.1-70B: 29.8 (#292)
| Benchmark | Gemma 4 31B IT | Llama 3.1-70B |
|---|---|---|
| WeirdML | 52.3% | 9% |
| LMArena Coding | 1459 | 1260 |
| LMArena WebDev | 1364 | — |
| SciCode | 43.4% | — |
| BigCodeBench Instruct | — | 46.1% |
| BigCodeBench Complete | — | 54.8% |
| ALE-Bench | 925.5 | — |
Agentic & Tool Use Not comparable
Gemma 4 31B IT: —, Llama 3.1-70B: 24.9 (#104)
| Benchmark | Gemma 4 31B IT | Llama 3.1-70B |
|---|---|---|
| TheAgentCompany | — | 6.9% |
| BALROG | — | 27.9% |
Reasoning Gemma 4 31B IT leads
Gemma 4 31B IT: 26.4 (#119), Llama 3.1-70B: 20.8 (#215)
| Benchmark | Gemma 4 31B IT | Llama 3.1-70B |
|---|---|---|
| LMArena Hard Prompts | 1448 | 1241 |
| DTBench | 82.7% | 60% |
| LMCA | 39.3% | 14.8% |
| Epoch Capabilities Index | 142.74 | 125.91 |
| Kagi LLM Benchmark | 63.5% | — |
| NYT Connections (extended) | 70.6% | — |
| CritPt | 1.4% | — |
| Chess Puzzles | 5% | — |
| Thematic Generalization | 53% | — |
| Surface Evolver Bench | 30.6% | — |
Math Gemma 4 31B IT leads
Gemma 4 31B IT: 41.3 (#72), Llama 3.1-70B: 12.7 (#298)
| Benchmark | Gemma 4 31B IT | Llama 3.1-70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.3% | 3.6% |
| LMArena Math | 1464 | 1252 |
| Omni-MATH | — | 21% |
| MATH Level 5 | — | 36.7% |
Knowledge Gemma 4 31B IT leads
Gemma 4 31B IT: 37.9 (#141), Llama 3.1-70B: 24.1 (#264)
| Benchmark | Gemma 4 31B IT | Llama 3.1-70B |
|---|---|---|
| GPQA Diamond | 75.8% | 44.2% |
| LMArena Expert | 1465 | 1209 |
| SimpleQA Verified | 10.4% | — |
| MMLU-Pro | — | 65.3% |
| Vectara Hallucination Rate | 7.4% | — |
| GPQA (HELM) | — | 42.6% |
| MMLU | — | 80.1% |
Multimodal Not comparable
Gemma 4 31B IT: 41.0 (#32), Llama 3.1-70B: —
| Benchmark | Gemma 4 31B IT | Llama 3.1-70B |
|---|---|---|
| LMArena Vision | 1278 | — |
| LMArena Document | 1425 | — |
Multilingual Gemma 4 31B IT leads
Gemma 4 31B IT: 54.0 (#52), Llama 3.1-70B: 38.9 (#220)
| Benchmark | Gemma 4 31B IT | Llama 3.1-70B |
|---|---|---|
| LMArena Non-English | 1431 | 1219 |
| LMArena Chinese | 1478 | 1215 |
| LMArena French | 1436 | 1260 |
| LMArena Russian | 1462 | 1234 |
| LMArena Spanish | 1443 | 1253 |
| LMArena German | — | 1222 |
| LMArena Japanese | — | 1132 |
| LMArena Korean | — | 1140 |
Instruction Following Gemma 4 31B IT leads
Gemma 4 31B IT: 75.6 (#58), Llama 3.1-70B: 65.5 (#218)
| Benchmark | Gemma 4 31B IT | Llama 3.1-70B |
|---|---|---|
| LMArena Instruction Following | 1433 | 1231 |
| IFEval | — | 82.1% |
Long Context Gemma 4 31B IT leads
Gemma 4 31B IT: 44.2 (#67), Llama 3.1-70B: 37.6 (#208)
| Benchmark | Gemma 4 31B IT | Llama 3.1-70B |
|---|---|---|
| LMArena Longer Query | 1446 | 1241 |
Writing & Preference Gemma 4 31B IT leads
Gemma 4 31B IT: 60.7 (#92), Llama 3.1-70B: 35.6 (#262)
| Benchmark | Gemma 4 31B IT | Llama 3.1-70B |
|---|---|---|
| LMArena Text | 1443 | 1261 |
| LMArena Creative Writing | 1416 | 1232 |
| EQ-Bench Creative Writing | 1368 | 784 |
| LMArena Multi-Turn | 1452 | 1256 |
| WildBench | — | 75.8% |
| EQ-Bench 4 | 1120 | — |
Frequently asked questions
Is Gemma 4 31B IT better than Llama 3.1-70B?
Gemma 4 31B IT is the stronger model overall, scoring 43.2 to 29.4 on the Noometry Index.
Which is cheaper, Gemma 4 31B IT or Llama 3.1-70B?
Gemma 4 31B IT is cheaper. It lists at $0.09 per million input tokens and $0.34 per million output tokens; Llama 3.1-70B lists at $0.40 and $0.40.
Is Gemma 4 31B IT or Llama 3.1-70B better for coding?
Gemma 4 31B IT scores higher on coding benchmarks: 41.9 versus 29.8 in the Noometry coding category.
Which has the bigger context window?
Gemma 4 31B IT does, with 262K tokens against 128K.
How many benchmarks do Gemma 4 31B IT and Llama 3.1-70B share?
21 benchmarks have published results for both models. Gemma 4 31B IT has 35 scored results on Noometry and Llama 3.1-70B has 35.