Model
Model explorer

Llama 3.1 8B

OPEN
Meta · Llama 3.1 family · released Jul 23, 2024

Upgraded 8B with 128K context, native tool calling, and eight-language multilingual support.

ReasoningCodingVisionFunction callingTool useAgentic
553.8
Elo · rank #287
Parameters
8.03B
Active params
8.03B (dense)
Context
128K tokens
Architecture
Dense decoder-only Transformer (GQA)
License
Llama 3.1 Community License Agreement
Languages
8+
API price (in/out)
No hosted API
Modalities
text
Benchmark results
Bar shows position within the tracked field; marker = field best
ARC-ChallengeReasoning83.4%#37
best: Llama 3.1 405B · 96.9%
Arena EloHuman preference1211#39
best: Claude Fable 5 · 1505
BIG-Bench HardReasoning30.1%#133
best: ERNIE 4.5 300B-A47B · 94.3%
BFCLAgents76.1%#16
best: SmolLM3 3B (No Thinking) · 92.3%
Gaokao MathMath30.4%#10
best: Doubao-Seed-1.6 · 96.0%
GPQA DiamondReasoning30.4%#259
best: GPT-5.6 · 94.6%
GSM8KMath84.5%#69
best: Llama 3.1 405B · 96.8%
HellaSwagReasoning80.2%#79
best: Claude 3 Opus · 95.4%
HumanEvalCoding72.6%#75
best: Claude Opus 4.5 · 99.4%
IFEvalReasoning80.4%#90
best: Gemma 4 26B A4B · 98.5%
KoMT-BenchHuman preference6.06#5
best: LG EXAONE 3.0 7.8B Instruct · 8.92
MATH-500Math51.9%#123
best: GPT-5 · 99.4%
MBPP+Coding72.8%#11
best: Llama 3.1 405B · 88.6%
MGSMMath68.9%#35
best: OpenAI o4-mini · 93.7%
MMLU (EU-21 languages)Knowledge57.6%#5
best: Llama 3.1 70B · 77.1%
MMLU-ProKnowledge48.3%#136
best: Claude Fable 5 · 91.5%
MMLUKnowledge69.4%#149
best: OpenAI o3 · 92.9%
TruthfulQAKnowledge54.0%#42
best: Phi-3.5-MoE (16x3.8B, 6.6B active) · 77.5%
WinoGrandeReasoning76.2%#55
best: PaLM 2 · 90.9%
Run it locally
VRAM @ Q4
5.5 GB
VRAM @ FP16
16 GB
Fits on (Q4)
RTX 3060 12GBRTX 4070 Ti 16GBRTX 3090 24GBRTX 4090 24GBRTX 5090 32GBM4 Pro 48GBM3 Max 128GBM3 Ultra 512GBA100 80GBH100 80GBH200 141GBB200 192GB
~100 tok/s on RTX 4090 (Q4, llama.cpp)
Quantizations
GGUF Q4 · AWQ · GPTQ · MLX
Llama 3.1 family
Elo progression across releases
Llama 3.1 405BJul 20241237.4Llama 3.1 70BJul 2024980.8Llama 3.1 8BJul 2024553.8
Resources
HFmeta-llama/Meta-Llama-3.1-8BGHmeta-llama/llama-modelsDOCSllama.com/docs/model-cards-and-prompt-formats/llama3_1APIAvailable via Meta API
API price weights · each benchmark row carries its own source badge (see methodology)