what/llm/can/i/run
models/meta-llama-llama-4-maverick-17b-128e-instruct

Llama 4 Maverick 17B 128E

Sources arena, livebench · Confidence B
params
400B
MoE total
bench sources
2
mid conf
context
max tokens
license
Llama 4
community

Model meta

canonical
meta-llama/Llama-4-Maverick-17B-128E-Instruct
parameters
400B
organization
Meta
license
Llama 4
context
downloads

Benchmark breakdown

LiveBench
instruction following79.6
coding51.1
language47.2
overall59.3

Hardware fit matrix

by gpu tierQ2_K
150.0 GB
Q4_K_M
200.0 GB
Q5_K_M
250.0 GB
Q6_K
300.0 GB
Q8_0
400.0 GB
FP16
800.0 GB
G8 GB tier
rtx 3050 · 4060
G12 GB tier
rtx 3060 · 4070
G16 GB tier
rtx 4080 · 4060 ti 16g
G24 GB tier
rtx 3090 · 4090
G32 GB tier
rtx 5090 · m3 max
G48 GB tier
a6000 · m3 max 64
G80 GB tier
h100 · m3 ultra 128
full fit · production speed tight fit · usabledoesn't fit
Get personalized ranking
Tell us your machine — we'll tell you if this is actually your best pick, or what's better.
Rank for my hardware