VISION & MULTIMODAL · 紅葉
Llama
2 models from Meta, living in Maple Highlands.
| Model | Context | Input | Output | Accepts |
|---|---|---|---|---|
| Muse Glimmer 30Bmeta/muse-glimmer-30b | 131K | $0.300 | $1.20 | text, images |
| Llama 4 Maverickmeta-llama/llama-4-maverick | 524K | $0.190 | $0.650 | text, images |
Questions and answers
Which Llama model has the largest context window?
Llama 4 Maverick (Meta), with 524K tokens.
Which Llama model is cheapest per input token?
Llama 4 Maverick, at $0.190 per million input tokens ($0.650 per million output tokens).
Which Llama models accept images?
Muse Glimmer 30B, Llama 4 Maverick (2 of 2).
Which Llama models support reasoning?
Muse Glimmer 30B (1 of 2).
Answers are computed from the 2026-10-24 catalog snapshot.