ModelPriceWatch.com
Last scan 2026-09-22 Models tracked 266 Providers 35 Cheapest paid Granite 4.0 H Micro $0.017/Mtok in Every price links to its source

Muse Glimmer 30B

by Fireworks · 30B parameters

Current new · 45d budget open weights cheap tier
Email me when Muse Glimmer 30B gets cheaper
Availability: This is Fireworks' price for hosting Meta's model, not a price from Meta. Meta does not sell Muse Glimmer per token at all — its own Model API rate card (ai.developer.meta.com/docs/pricing-rate-limits, read 2026-08-28) prices only muse-spark-1.1/1.2 and muse-spark-1.2-contributor, and ships Glimmer as open weights under Apache 2.0. Together publishes the same $0.35/$1.50 for it. Fireworks' card carries no context column, so the 131,072-token window is the one Together's own card publishes for the same weights and the one Fireworks' OpenRouter endpoint reports.

Today's price · per 1M tokens

Input

$0.350

per 1M tokens

Output

$1.50

per 1M tokens

Blended

$0.637

blended $/1M — 3:1 weighted input:output

Cached input

$0.040

11% of input — prompt caching

Source: official Fireworks pricing · read MODELPRICEWATCH.COM · 2026-09-22

Price receipt

We read Fireworks's own pricing page on and found Muse Glimmer 30B listed with a price on the same row — that read is where the number above comes from, recorded as a new model. We keep the page text we read; its content hash is 57dbe7dd60.

Muse Glimmer 30B is available on 2 hosts. See the full cheapest-first comparison on the consolidated Muse Glimmer 30B page →

Overview

Meta's open-weights Muse Glimmer 30B on Fireworks. $0.35/$1.50 per 1M; cached input $0.04.

Run it yourself

Deploy this open model on rented GPUs

Open weights mean you can self-host instead of paying per-token API prices. These platforms let you serve it on demand — often cheaper at scale.

RunPod →
Rent GPUs by the second and serve any open-weight model yourself.
Novita AI →
Serverless open-model inference billed per token — no GPU ops.
Vast.ai →
GPU marketplace — rent spot/on-demand GPUs cheaply and serve any open-weight model.

Partner links — we may earn a commission if you sign up, at no cost to you. We only list platforms we'd recommend regardless.

Capabilities

struck through = not supported
Input 1/4
Text ✓ Image Audio Video
Output
Text ✓
Features 3/9
Prompt caching ✓ Reasoning Coding Fast inference Long context ✓ Open weights ✓ Multimodal Web search Realtime

Benchmark performance

accuracy % · higher is better
Avg benchmark score37%
Perf per $/Mtok58
GPQA Diamond vendor
83.5%
SWE-Bench Verified vendor
76%
Terminal-Bench 2.1
52%
MCP Atlas vendor
75.5%
OSWorld-Verified vendor
65.9%
Humanity's Last Exam
22%
Independent composite scores
  • AA Intelligence Index: 18.1 (Artificial Analysis composite across reasoning, knowledge and coding evals)
  • AA-Omniscience Index: -32.9 (−100–100)
  • GDPval-AA v2: 955.3 ELO (Real-world work tasks, human baseline = 1000)
  • LMSYS Chatbot Arena: 1427.2 ELO (Human preference)
How it stacks up
  • Ranks #80 of 84 benchmarked models by average score
  • Ranks #129 of 181 comparably-measured models by percentile score, across 6 independent measurements
  • Ranks #66 of 181 comparably-tested models by normalized performance per dollar
  • Strongest at Humanity's Last Exam — 22%, #44 of 65

60.9 tokens/sec output

Source: Artificial Analysis, Artificial Analysis (Terminus 2, high), Artificial Analysis (text-only, no tools, high) · updated · See full rankings →

Specifications

Provider
Fireworks
Context window
131K tokens
Modality
text
Parameters
30B
Open source
Yes — open weights available
Released
Status
Current
Last updated
Tags
open-weightsbudgetcaching

Availability verified: — listed on Fireworks's own page

Try on Fireworks Official pricing page