Llama-4-Scout-17B-16E-Instruct
Meta Llama- Arena score
- 1279
- Among open models
- #94of 148
- Among all on Arena
- #278
- Votes
- 29,775
Measured on Llama 4 Scout 17B 16E Instruct, the served version, not a local build. LMArena, CC BY 4.0, published Nov 11, 2026.
Score by category
CategoryScoreOpen rankArena rankVotes
- English1304#91#26613K
- Hard prompts, English1293#89#2615.7K
- Math1287#84#2491.9K
- Applied math1286#85#2541.8K
- Coding1286#93#2765.2K
- Software & IT1286#96#2769.5K
- Legal & government1285#92#2672K
- French1282#70#226427
- Multi-turn1280#88#2675.2K
- Science1279#95#2785K
- Spanish1278#76#230709
- Medicine & health1276#91#2641.8K
- German1272#70#220800
- Writing & literature1268#88#2726.7K
- Hard prompts1266#97#27812.9K
- Long prompts1265#95#2815.8K
- Polish1265#70#2202.5K
- Russian1262#89#2692.3K
- Business & finance1257#93#2704.9K
- Chinese1255#89#2692.1K
- Non-English1252#94#28116.8K
- Creative writing1249#92#2733.8K
- Instructions1248#95#2847.5K
- Media & sports1242#91#2735.3K
- Expert1235#96#2721.5K
- Excluding ties1208#94#27621.4K
- Japanese1207#61#214764
- Korean1207#67#218651
About as strong, smaller
Compare
granite-4.2-3bIBM Granite12923.7B
gemma-3-4b-itGoogle12914.3B

gemma-3n-E4B-itGoogle13067.8B

Molmo2-8BAi212878.7B
Built on it
Ways to run it
- GGUF · llama.cpp, Ollama, LM Studiounsloth/Llama-4-Scout-17B-16E-Instruct-GGUF22 builds
- MLX · Apple silicon: mlx-lm, LM Studiolmstudio-community/Llama-4-Scout-17B-16E-MLX-text-8bit3 builds
- AWQ · vLLM, SGLangkishizaki-sci/Llama-4-Scout-17B-16E-Instruct-AWQ1 build
Most downloaded
- Llama-4-Scout-17B-16E-Instruct-GGUFunsloth · Quantization38K downloads
- Llama-4-Scout-17B-16E-InstructRedHatAI · Fine-tune22.3K downloads
- Llama-4-Scout-17B-16E-Instruct-quantized.w4a16RedHatAI · Quantization9K downloads
- Llama-4-Scout-17B-16E-Instructunsloth · Fine-tune6.4K downloads
- meta-llama_Llama-4-Scout-17B-16E-Instruct-old-GGUFbartowski · Quantization5.1K downloads
BenchClub runs
Not runnable yet
Runs open once a Llama-4-Scout-17B-16E-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 80–300B models
See more
DeepSeek-V4-FlashDeepSeek1.2M291B

MiniMax-M2.5MiniMax1M229B
Inkling-SmallThinking Machines1M266B
Hy3Tencent987.2K299B