Meta: Muse Glimmer 30B.
A dense 30B open-weights model from Meta that takes text and images and is tuned for local agent work, with a 4-bit build that fits in 24 to 32 GB of VRAM.
1× Nvidia RTX A4500 · $0.19/hr.
Most-aggressive quantisation we have a working recommendation for. Lower precision = less VRAM = cheaper hardware, at a small accuracy cost.
Cheapest hosted endpoints.
Cheapest hardware per quantisation.
Each row is one quantisation tier (the same weights compressed differently). Lower precision → lower VRAM → cheaper hardware, at the cost of small accuracy loss. $/hr refreshed hourly from each provider's API.
| Quantisation | Cheapest GPU config | Total VRAM | Live $/hr | tokens/sec | |
|---|---|---|---|---|---|
|
FP16
FP16 — half precision (default)
|
80 GB | $1.09/hr | — | Compare → | |
|
FP8
FP8 — 8-bit float (Hopper / Blackwell)
|
40 GB | $1.39/hr | — | Compare → | |
|
INT4
INT4 — 4-bit integer (~4× VRAM saving)
|
20 GB | $0.19/hr | — | Compare → |
What it costs per month across providers.
Estimate your monthly bill for Meta: Muse Glimmer 30B across every host that publishes per-token pricing. Slide your token volumes; the chart + table re-rank cheapest-first.
Cheapest provider on the left.
Total monthly cost — input + output tokens combined.
Bill breakdown.
Rent the GPU instead of paying per token.
For an open-weights model like Meta: Muse Glimmer 30B, you can rent a GPU and serve inference yourself. The math: cheapest GPU rental × 730 hours/month + your electricity rate × power draw.
Assumes the GPU runs 24/7 at ~85% utilisation. If your traffic is bursty, you'll pay less for the API and probably more for the GPU (idle hours still cost rental). The breakeven analysis lives on the Self-host vs API breakeven tool.
About Meta: Muse Glimmer 30B.
Muse Glimmer is a dense 30B-class model (about 29.6B parameters including its vision encoder) from Meta Superintelligence Lab, released with open weights under Apache 2.0. It reads text and images, outputs text, and handles over 131,000 tokens of context. It is built for agent tasks that run locally: tool calls, multi-step reasoning, recovering from failed calls and reading screenshots or documents. Meta compresses it to roughly 4-bit so it fits in 24 to 32 GB of VRAM, and ships a small drafter model for faster generation.