Meta-Llama-3-8B-Instruct
Meta Llama- Arena score
- 1166
- Among open models
- #118of 148
- Among all on Arena
- #340
- Votes
- 104,642
Measured on Llama 3 8B Instruct, the served version, not a local build. LMArena, CC BY 4.0, published Nov 11, 2026.
Score by category
CategoryScoreOpen rankArena rankVotes
- English1213#114#32961.7K
- Spanish1173#90#2691.7K
- Science1170#114#33417.3K
- Legal & government1168#115#3315.6K
- Hard prompts, English1164#116#33819.9K
- French1159#91#2721.6K
- Software & IT1156#119#34229.6K
- Applied math1153#117#33512.8K
- Multi-turn1152#117#33814.8K
- Coding1151#121#34618.4K
- Math1151#118#33614.3K
- Creative writing1150#114#33615.4K
- Medicine & health1148#111#3235.4K
- Writing & literature1148#119#34325.5K
- Polish1137#80#241257
- Hard prompts1133#122#34730.2K
- Media & sports1132#116#33915.4K
- Long prompts1127#123#35011.2K
- Instructions1127#121#34537.7K
- Business & finance1127#118#34110.8K
- Expert1113#118#3325.4K
- Russian1109#121#34610K
- German1104#95#2893.4K
- Non-English1098#123#35242.9K
- Chinese1076#123#35110.8K
- Excluding ties1030#118#34267.7K
- Korean1004#89#2762.4K
- Japanese968#87#2782K
| Model | Params | Score |
|---|---|---|
| 8.0B | — | |
| 8.0B | 1166 | |
| 70.6B | 1221 |
About as strong, smaller
Compare
gemma-2-2b-itGoogle11562.6B
granite-4.2-3bIBM Granite12923.7B
Built on it
Top 10 among open models in#1Merges
Ways to run it
- GGUF · llama.cpp, Ollama, LM StudioMaziyarPanahi/Meta-Llama-3-8B-Instruct-GGUF164 builds
- MLX · Apple silicon: mlx-lm, LM Studioweber50432/lora-Meta-Llama-3-8B-Instruct1 build
- AWQ · vLLM, SGLangsolidrust/Meta-Llama-3-8B-Instruct-AWQ5 builds
- GPTQ · vLLM, SGLangMaziyarPanahi/Meta-Llama-3-8B-Instruct-GPTQ11 builds
Most downloaded
- Meta-Llama-3-8B-Instruct-GGUFMaziyarPanahi · Quantization60K downloads
- chronicle-narrator-ggufGWsuryaYT · Quantization44.3K downloads
- Meta-Llama-3-8B-Instruct-GGUFQuantFactory · Quantization26.2K downloads
- Meta-Llama-3-8B-Instruct-llamafilemozilla-ai · Fine-tune17.9K downloads
- Llama3-8B-Chinese-Chatshenzhi-wang · Fine-tune9.4K downloads
BenchClub runs
Not runnable yet
Runs open once a Meta-Llama-3-8B-Instruct build is registered.
More from Meta Llama
See more
Llama-3.1-8B-Instruct4.2M8.0B

Llama-3.2-1B-Instruct1.3M1.2B

Llama-3.2-3B-Instruct831.4K3.2B

Llama-3.2-1B644.1K1.2B
Other 3–9B models
See more
NVIDIA-Nemotron-3-Nano-4B-BF16NVIDIA1.9M4.0B
granite-4.1-8bIBM Granite79.4K8.8B
Molmo2-8BAi247.6K8.7B
Spark-X2.6-4BXHToken32.9K4.1B