Llama-4-Maverick-17B-128E-Instruct
Meta Llama- Arena score
- 1287
- Among open models
- #90of 148
- Among all on Arena
- #265
- Votes
- 39,386
Measured on Llama 4 Maverick 17B 128E Instruct, the served version, not a local build. LMArena, CC BY 4.0, published Nov 11, 2026.
Score by category
CategoryScoreOpen rankArena rankVotes
- English1302#94#27018.9K
- Coding1301#88#2626.9K
- Math1300#82#2442.8K
- Software & IT1299#92#26612.3K
- Hard prompts, English1295#88#2597.7K
- Applied math1295#82#2472.6K
- Spanish1293#73#220816
- German1290#66#2071K
- Multi-turn1289#85#2586.7K
- Science1288#93#2716.7K
- Russian1286#82#2463.1K
- Legal & government1286#91#2662.6K
- Hard prompts1281#89#26415.8K
- Long prompts1280#88#2707.1K
- Chinese1278#86#2572.6K
- Writing & literature1277#84#2628.7K
- Medicine & health1276#90#2632.3K
- Polish1275#67#2162.7K
- Non-English1269#87#26020.5K
- Instructions1267#88#26310.5K
- Creative writing1266#84#2595.1K
- Business & finance1261#90#2676.1K
- French1260#77#243594
- Expert1258#87#2532K
- Media & sports1250#88#2666.9K
- Excluding ties1221#90#26428.1K
- Japanese1206#62#215936
- Korean1204#69#221798
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-Maverick-17B-128E-Instruct-GGUF8 builds
- MLX · Apple silicon: mlx-lm, LM Studiobaa-ai/Llama-4-Maverick-17B-128E-Instruct-RAM-170GB-MLX3 builds
- AWQ · vLLM, SGLangkishizaki-sci/Llama-4-Maverick-17B-128E-Instruct-AWQ1 build
Most downloaded
- Llama-4-Maverick-17B-128E-Instruct-GGUFunsloth · Quantization29.6K downloads
- Llama-4-Maverick-17B-128E-Instruct-quantized.w4a16RedHatAI · Quantization16.3K downloads
- Llama-4-Maverick-17B-128E-Instructunsloth · Fine-tune6.4K downloads
- Llama-4-Maverick-17B-128E-Instruct-GGUF-UDunsloth · Quantization1.7K downloads
- meta-llama.Llama-4-Maverick-17B-128E-Instruct-GGUFDevQuasar-12 · Quantization763 downloads
BenchClub runs
Not runnable yet
Runs open once a Llama-4-Maverick-17B-128E-Instruct build is registered.
More from Meta Llama
See more
Llama-3.1-8B-Instruct4.1M8.0B

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

Llama-3.2-3B-Instruct823.8K3.2B

Llama-3.2-1B635.8K1.2B
Other >300B models
See more
DeepSeek-V3.2DeepSeek9.9M685B
GLM-5.3-FlashZ.ai6.4M321B
InklingThinking Machines1.4M952B
NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16NVIDIA1M561B