# GPT-4.1 vs Llama 3-8B

> GPT-4.1 is the stronger model overall, scoring 35.9 to 25.5 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-4-1-vs-llama-3-8b
- Last updated: 2026-10-10
- Shared benchmarks: 24

## Summary

- They share 24 benchmarks with published results for both. GPT-4.1 scores higher in 7 categories and Llama 3-8B in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where GPT-4.1 leads 37.1 to 7.8.
- The biggest single-benchmark swing is MATH Level 5: 83% for GPT-4.1 and 6.1% for Llama 3-8B.
- Llama 3-8B has downloadable open weights; the other is API-only.

## Snapshot

| | GPT-4.1 | Llama 3-8B |
|---|---|---|
| Provider | OpenAI | Meta |
| Noometry Index | 35.9 | 25.5 |
| Rank | 219 | 344 |
| Context | 1.05M | — |
| Input $/M | $2 | — |
| Output $/M | $8 | — |
| Weights | Proprietary | Open |

## Coding

- GPT-4.1: 34.4 (#238)
- Llama 3-8B: 31.0 (#289)

| Benchmark | GPT-4.1 | Llama 3-8B |
|---|---|---|
| LMArena Coding | 1391 | 1152 |
| SWE-bench Verified | 48.5% | — |
| SWE-bench Verified (bash only) | 39.6% | — |
| Aider Polyglot | 52.4% | — |
| WeirdML | 39% | — |
| BigCodeBench Instruct | — | 31.9% |
| BigCodeBench Complete | — | 36.9% |
| CadEval | 42% | — |
| ALE-Bench | 558.1 | — |
| HumanEval+ | — | 56.7% |
| MBPP+ | — | 54.8% |

## Agentic & Tool Use

- GPT-4.1: 34.7 (#43)
- Llama 3-8B: —

| Benchmark | GPT-4.1 | Llama 3-8B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 54% | — |

## Reasoning

- GPT-4.1: 11.7 (#339)
- Llama 3-8B: 14.3 (#326)

| Benchmark | GPT-4.1 | Llama 3-8B |
|---|---|---|
| Chess Puzzles | 6% | 0% |
| LMArena Hard Prompts | 1384 | 1133 |
| DTBench | 68.3% | 43.9% |
| Epoch Capabilities Index | 136.78 | 116.45 |
| ForecastBench | 61.5 | 58.6 |
| ARC-AGI-2 | 0.4% | — |
| SimpleBench | 27% | — |
| Kagi LLM Benchmark | 52.3% | — |
| ARC-AGI-1 | 5.5% | — |
| EnigmaEval | 2.2% | — |
| LMCA | 25.6% | — |
| Adversarial NLI | — | 57.3% |
| WinoGrande | — | 75.7% |

## Math

- GPT-4.1: 22.3 (#280)
- Llama 3-8B: 8.8 (#323)

| Benchmark | GPT-4.1 | Llama 3-8B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 38.3% | 1.9% |
| LMArena Math | 1370 | 1151 |
| MATH Level 5 | 83% | 6.1% |
| FrontierMath (Tiers 1-3) | 6% | — |
| Omni-MATH | 47.1% | — |
| FrontierMath (Feb 2025 set) | 5.5% | — |
| FrontierMath Tier 4 (v1) | 0% | — |

## Knowledge

- GPT-4.1: 37.1 (#160)
- Llama 3-8B: 7.8 (#308)

| Benchmark | GPT-4.1 | Llama 3-8B |
|---|---|---|
| GPQA Diamond | 66.9% | 26.1% |
| LMArena Expert | 1364 | 1113 |
| Humanity's Last Exam | 5.4% | — |
| SimpleQA Verified | 31.1% | — |
| MMLU-Pro | 81.1% | — |
| Vectara Hallucination Rate | 5.6% | — |
| GPQA (HELM) | 65.9% | — |
| ARC (AI2) Challenge | — | 82.8% |
| MMLU | — | 68.8% |
| OpenBookQA | — | 82.6% |
| TriviaQA | — | 67.7% |

## Multimodal

- GPT-4.1: 38.2 (#67)
- Llama 3-8B: —

| Benchmark | GPT-4.1 | Llama 3-8B |
|---|---|---|
| LMArena Vision | 1211 | — |
| GeoBench | 72% | — |

## Multilingual

- GPT-4.1: 49.4 (#133)
- Llama 3-8B: 30.8 (#261)

| Benchmark | GPT-4.1 | Llama 3-8B |
|---|---|---|
| LMArena Non-English | 1370 | 1098 |
| LMArena Chinese | 1382 | 1076 |
| LMArena French | 1382 | 1159 |
| LMArena German | 1381 | 1104 |
| LMArena Japanese | 1319 | 967 |
| LMArena Korean | 1339 | 1004 |
| LMArena Russian | 1377 | 1109 |
| LMArena Spanish | 1376 | 1173 |

## Instruction Following

- GPT-4.1: 71.3 (#153)
- Llama 3-8B: 58.4 (#260)

| Benchmark | GPT-4.1 | Llama 3-8B |
|---|---|---|
| LMArena Instruction Following | 1367 | 1127 |
| IFEval | 83.8% | — |

## Long Context

- GPT-4.1: 40.0 (#163)
- Llama 3-8B: 34.2 (#251)

| Benchmark | GPT-4.1 | Llama 3-8B |
|---|---|---|
| LMArena Longer Query | 1385 | 1128 |
| Fiction.LiveBench | 63.9% | — |

## Writing & Preference

- GPT-4.1: 57.6 (#125)
- Llama 3-8B: 37.5 (#256)

| Benchmark | GPT-4.1 | Llama 3-8B |
|---|---|---|
| LMArena Text | 1383 | 1166 |
| LMArena Creative Writing | 1363 | 1150 |
| LMArena Multi-Turn | 1398 | 1152 |
| EQ-Bench Creative Writing | 1420 | — |
| WildBench | 85.4% | — |

## FAQ

### Is GPT-4.1 better than Llama 3-8B?

GPT-4.1 is the stronger model overall, scoring 35.9 to 25.5 on the Noometry Index.

### Is GPT-4.1 or Llama 3-8B better for coding?

GPT-4.1 scores higher on coding benchmarks: 34.4 versus 31.0 in the Noometry coding category.

### How many benchmarks do GPT-4.1 and Llama 3-8B share?

24 benchmarks have published results for both models. GPT-4.1 has 52 scored results on Noometry and Llama 3-8B has 34.