Model comparison
Gemma 4 31B IT vs Llama 3.1-8B
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 23.0 on the Noometry Index. Llama 3.1-8B costs 2.7× less per token, which makes it the better buy when Gemma 4 31B IT's lead doesn't matter for your workload.
Last verified . 24 shared benchmarks.
Google
MetaSummary
- They share 24 benchmarks with published results for both. Gemma 4 31B IT scores higher in 8 categories and Llama 3.1-8B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Gemma 4 31B IT leads 43.2 to 10.2.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 73.3% for Gemma 4 31B IT and 1.7% for Llama 3.1-8B.
- Llama 3.1-8B is cheaper at $0.05 / $0.08 per million input/output tokens, against $0.09 / $0.34 for Gemma 4 31B IT.
- Gemma 4 31B IT accepts more context: 262K tokens versus 128K.
Side by side
| Gemma 4 31B IT | Llama 3.1-8B | |
|---|---|---|
| Provider | Meta | |
| Noometry Index | 43.5 | 23.0 |
| Released | 2026-04-02 | 2024-07-23 |
| Weights | Open | Open |
| Context window | 262K | 128K |
| Max output | 33K | 4K |
| Input $ / M tokens | $0.09 | $0.05 |
| Output $ / M tokens | $0.34 | $0.08 |
| Results tracked | 35 | 43 |
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: 42.3 (#108), Llama 3.1-8B: 20.2 (#340)
| Benchmark | Gemma 4 31B IT | Llama 3.1-8B |
|---|---|---|
| SciCode | 43.4% | 13.2% |
| WeirdML | 52.3% | 1.7% |
| LMArena Coding | 1459 | 1195 |
| LMArena WebDev | 1366 | — |
| BigCodeBench Instruct | — | 32.8% |
| BigCodeBench Complete | — | 40.5% |
| ALE-Bench | 925.5 | — |
| HumanEval+ | — | 62.8% |
| MBPP+ | — | 55.6% |
Agentic & Tool Use Not comparable
Gemma 4 31B IT: —, Llama 3.1-8B: 22.5 (#131)
| Benchmark | Gemma 4 31B IT | Llama 3.1-8B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 25.8% |
| BALROG | — | 15.1% |
Reasoning Gemma 4 31B IT leads
Gemma 4 31B IT: 27.2 (#122), Llama 3.1-8B: 14.9 (#321)
| Benchmark | Gemma 4 31B IT | Llama 3.1-8B |
|---|---|---|
| CritPt | 1.4% | 0% |
| Chess Puzzles | 5% | 0% |
| LMArena Hard Prompts | 1448 | 1175 |
| DTBench | 82.7% | 50.9% |
| LMCA | 39.3% | 5.4% |
| Epoch Capabilities Index | 142.74 | 116.57 |
| Kagi LLM Benchmark | 63.5% | — |
| NYT Connections (extended) | 70.6% | — |
| Thematic Generalization | 53% | — |
| Surface Evolver Bench | 30.6% | — |
| PIQA | — | 81.2% |
Math Gemma 4 31B IT leads
Gemma 4 31B IT: 43.2 (#81), Llama 3.1-8B: 10.2 (#317)
| Benchmark | Gemma 4 31B IT | Llama 3.1-8B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.3% | 1.7% |
| LMArena Math | 1465 | 1179 |
| Omni-MATH | — | 13.7% |
| MATH Level 5 | — | 22.9% |
| GSM8K | — | 82.4% |
Knowledge Gemma 4 31B IT leads
Gemma 4 31B IT: 37.9 (#151), Llama 3.1-8B: 8.0 (#307)
| Benchmark | Gemma 4 31B IT | Llama 3.1-8B |
|---|---|---|
| GPQA Diamond | 75.8% | 27% |
| LMArena Expert | 1465 | 1144 |
| SimpleQA Verified | 10.4% | — |
| MMLU-Pro | — | 40.6% |
| Vectara Hallucination Rate | 7.4% | — |
| GPQA (HELM) | — | 24.7% |
| BoolQ | — | 82.8% |
| MMLU | — | 56.1% |
Multimodal Not comparable
Gemma 4 31B IT: 41.6 (#34), Llama 3.1-8B: —
| Benchmark | Gemma 4 31B IT | Llama 3.1-8B |
|---|---|---|
| LMArena Vision | 1277 | — |
| LMArena Document | 1425 | — |
Multilingual Gemma 4 31B IT leads
Gemma 4 31B IT: 53.8 (#57), Llama 3.1-8B: 34.0 (#249)
| Benchmark | Gemma 4 31B IT | Llama 3.1-8B |
|---|---|---|
| LMArena Non-English | 1431 | 1148 |
| LMArena Chinese | 1476 | 1151 |
| LMArena French | 1435 | 1177 |
| LMArena Russian | 1460 | 1158 |
| LMArena Spanish | 1444 | 1169 |
| LMArena German | — | 1144 |
| LMArena Japanese | — | 1061 |
| LMArena Korean | — | 1053 |
Instruction Following Gemma 4 31B IT leads
Gemma 4 31B IT: 75.5 (#61), Llama 3.1-8B: 58.9 (#258)
| Benchmark | Gemma 4 31B IT | Llama 3.1-8B |
|---|---|---|
| LMArena Instruction Following | 1433 | 1159 |
| IFEval | — | 74.3% |
Long Context Gemma 4 31B IT leads
Gemma 4 31B IT: 44.2 (#71), Llama 3.1-8B: 35.8 (#238)
| Benchmark | Gemma 4 31B IT | Llama 3.1-8B |
|---|---|---|
| LMArena Longer Query | 1446 | 1182 |
Writing & Preference Gemma 4 31B IT leads
Gemma 4 31B IT: 60.5 (#96), Llama 3.1-8B: 29.7 (#290)
| Benchmark | Gemma 4 31B IT | Llama 3.1-8B |
|---|---|---|
| LMArena Text | 1443 | 1187 |
| LMArena Creative Writing | 1415 | 1154 |
| EQ-Bench Creative Writing | 1368 | 713 |
| LMArena Multi-Turn | 1452 | 1172 |
| WildBench | — | 68.7% |
| EQ-Bench 4 | 1120 | — |
Frequently asked questions
Is Gemma 4 31B IT better than Llama 3.1-8B?
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 23.0 on the Noometry Index. Llama 3.1-8B costs 2.7× less per token, which makes it the better buy when Gemma 4 31B IT's lead doesn't matter for your workload.
Which is cheaper, Gemma 4 31B IT or Llama 3.1-8B?
Llama 3.1-8B is cheaper. It lists at $0.05 per million input tokens and $0.08 per million output tokens; Gemma 4 31B IT lists at $0.09 and $0.34.
Is Gemma 4 31B IT or Llama 3.1-8B better for coding?
Gemma 4 31B IT scores higher on coding benchmarks: 42.3 versus 20.2 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 3.1-8B share?
24 benchmarks have published results for both models. Gemma 4 31B IT has 35 scored results on Noometry and Llama 3.1-8B has 43.