GeekRadar / The comparison desk

COMPARE.

AI models

Muse Spark Open

AI model Released Oct 15, 2026

Context window: 1,048,576 tokens
Maximum output: 131,072 tokens
Open weights: Yes

Capability & preference

General capability · ECI

A cross-benchmark capability estimate. Index points, not a percentage; intervals show uncertainty.

Muse Spark Open 155.1^* Muse Spark Open · 95% interval 153.1–157.3

Model performance

Radar task score: four tests, equal weight. An experimental summary of these tasks, not a universal intelligence rating.

Muse Spark Open

Tested configuration Radar task score

Unrated

3/4 core tests · not enough evidence

5 benchmark results for this configuration

Scientific reasoning

GPQA Diamond · 25%

Expert-written science questions; strict multiple-choice scoring. Chance accuracy is 25%.

Muse Spark Open 88.6^*% Approx. 95% interval: 84.9–92.2% · 2026-10-15

Novel problem solving

Mystery Game Puzzles · 25%

Best-next-move puzzles in an undisclosed game variant. A narrow reasoning test, not general intelligence.

Muse Spark Open Not available

Advanced mathematics

FrontierMath Tiers 1-3 · 25%

Private v2 mathematics problems with Python tools. Earlier versions and Tier 4 are excluded.

Muse Spark Open 69.8^*% Approx. 95% interval: 64.5–75.2% · 2026-10-15

Factual answers

SimpleQA Verified · 25%

Correct factual answers under a guessing prompt, graded by a model. Not a hallucination or abstention score.

Muse Spark Open 56.6^*% Approx. 95% interval: 53.5–59.7% · 2026-10-15

Competition mathematics

OTIS Mock AIME 2024-2025 · Separate result

Competition problems with model-extracted answers checked against an exact key. Earlier grading procedures are excluded.

Muse Spark Open 98.1^*% Approx. 95% interval: 95.6–100.0% · 2026-10-15

Research mathematics

FrontierMath Tier 4 v2 · Separate result

The hardest private FrontierMath problems, corrected v2 set, with Python tools. Separate from Tiers 1-3 in the core score.

Muse Spark Open 31.7^*% Approx. 95% interval: 17.3–46.1% · 2026-10-15

Bars show reported task scores. Whiskers are approximate 95% intervals where standard errors are available. Different reasoning budgets are not equal-cost tests. Missing results are never zero.

Run it locally

Inference · memory estimate

Muse Spark Open

Not enough verified architecture information for a reliable VRAM estimate.

API cost

USD / month

Input tokens / monthOutput tokens / month

Standard text-token estimate. Excludes taxes, caching, batch discounts, reasoning and multimodal surcharges. Provider limits and regional availability may differ.

Muse Spark Open

Provider offer

Muse Spark Open / Pricing tiers 2

Provider / scope	Rate	Currency	Amount / unit
Fireworks AIMuse Spark OpenPriority	Cached input	USD	0.225 / 1M tokens
	Input	USD	1.875 / 1M tokens
	Output	USD	6.375 / 1M tokens
Fireworks AIMuse Spark OpenStandard	Cached input	USD	0.15 / 1M tokens
	Input	USD	1.25 / 1M tokens
	Output	USD	4.25 / 1M tokens

Capabilities

Context window: 1,048,576 tokens
Maximum output: 131,072 tokens
Reasoning: Yes
Tool calling: Yes
Open weights: Yes
Knowledge cutoff: 2026-01-04
Model license: Muse Model License

Input modalities

text
image
video
audio

Output modalities

text

* Provisional data