Skip to content

meta/muse-glimmer-30b pricing

Chat

meta/muse-glimmer-30b · Vercel AI Gateway · context —

Price per 1M tokens

In effect since Oct 15, 2026

meta/muse-glimmer-30b token prices
Input (uncached)$0.35
OutputIncludes reasoning tokens$1.50
Cache read$0.04
  • No long-context tier: one rate for every prompt size.

Price history

1 version since Oct 15, 2026

  • Input
  • Output
Show data as table
meta/muse-glimmer-30b price history per 1M tokens
Effective fromInput / 1MOutput / 1M
Oct 15, 2026$0.35$1.50
Oct 16, 2026$0.35$1.50

Cost calculator

Calculated with the same function that AI Cost Map uses for your real events.

Includes reasoning tokens.

Read from the prompt cache.

Cost per call

$0.0014

Cost per month

$145.00

Similar models

Same model type, nearest blended price (3 input : 1 output).

Similar models compared with meta/muse-glimmer-30b
ModelInputOutputContextBlended vs this
meta/muse-glimmer-30b (this model)
Vercel AI Gateway
$0.35$1.50——
meta-models/muse-glimmer-30b
Together AI
$0.35$1.50—× 1.0
llama-3.3-70b-versatile
Groq
$0.59$0.79—× 1.0
meta/llama-3-70b
Vercel AI Gateway
$0.59$0.79—× 1.0
qwen/qwen3-vl-235b-a22b-instruct
OpenRouter
$0.21$1.90—× 0.99
baidu/ernie-4.5-vl-424b-a47b
Novita AI
$0.42$1.25—× 0.98
moonshot-v1-8k
Moonshot AI
$0.20$2.00—× 1.0