Up to four models
Gemma 3 4B vs Llama 3.1-8B
On the index (GPT-4 = 100), Gemma 3 4B scores 89 and Llama 3.1-8B 89. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
8962–117
WeightsOpen
ReleasedMar 2025
Price in / out$0.050 / $0.10
Context131K
Core tests taken1 of 9
8970–107
WeightsOpen
ReleasedJul 2024
Price in / out$0.020 / $0.040
Context131K
Core tests taken1 of 9
Gemma 3 4B and Llama 3.1-8B are too close to call (1 point apart, within the uncertainty), ahead on 0 of 1 shared core tests, at 2.5× the price.
Test by testthird-party results
Every result as a table
| Test | Gemma 3 4B | Llama 3.1-8B |
|---|---|---|
| Reasoning & knowledge | ||
| GPQA Diamond (Epoch AI) | 23.2% | 27.0% |
| Maths | ||
| AIME-style maths (OTIS mock) | 7.5% | 1.7% |
| Agents | ||
| DTBench | 50.9% | 50.9% |
| LMCA | 2.8% | 5.4% |
Tests at least two of these models have taken. Bold rows are the core tests.