llama-4-maverick
llama-4-maverickSamples: 0
Meta Llama 4 Maverick 17B/128E mixture-of-experts, natively multimodal with a 1M token context window for chat, vision and tool use.
- Input / 1M tokens
- $0.35
- Output / 1M tokens
- $1.15
- Context window
- 524,288
- Route
- originPlus routes
Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.
Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.
EQ-Bench 3
meta-llama/Llama-4-Maverick-17B-128E-Instruct: 864.8
83 Models · 2026-10-25
Creative Writing v3
meta-llama/Llama-4-Maverick-17B-128E-Instruct: 858
145 Models · 2026-10-25
WritingBench · Literature & arts
llama-4-maverick-17b-128e: 51.25
54 Models · 2026-10-25
WritingBench · Character design
llama-4-maverick-17b-128e: 54.8
54 Models · 2026-10-25
WritingBench · Plot development
llama-4-maverick-17b-128e: 51.45
54 Models · 2026-10-25
WritingBench · Fan fiction
llama-4-maverick-17b-128e: 51.5
54 Models · 2026-10-25
Routes
Ability score divided by token cost, ranked across measured model routes. Cost uses equal input and output token counts; higher is better value.
| Use this model | ||||||
|---|---|---|---|---|---|---|
| origin | $0.35 | $1.15 | $0.35 | 4.4/10 | 7.6/10 | Plans and tiers |
Quick start
Access and billingcurl --fail-with-body 'https://api.tokenreply.com/v1/chat/completions' \
-H "Authorization: Bearer $TOKENREPLY_API_KEY" \
-H 'Content-Type: application/json' \
--data '{
"model": "llama-4-maverick",
"messages": [
{
"role": "user",
"content": "Reply with hello."
}
]
}'