Muse Spark 1.2 cost & VRAM calculator
What does Muse Spark 1.2 cost to run for your workload, and can you run it on your own hardware? Set your workload below - we compute per-provider API cost live and tell you honestly whether local hardware can run it.
This is where owning stops making sense
Muse Spark 1.2 is a large-billion-parameter model. No rig an individual can buy runs it - so unlike a workstation GPU model, there's no break-even to compute. The honest answer for nearly everyone is the per-provider API cost below.
Proprietary dense coding model from Meta, released August 5, 2026 alongside the Muse Code terminal agent. Native 1M-token context. Mandatory reasoning with five effort levels ( minimal, low, medium, high, xhigh; medium default). Multimodal: text, image, video, audio, and PDF input; text output. Pricing (Meta Model API standard): $1.25/1M input, $4.25/1M output; cached input $0.15/1M. Pricing (Contributor tier): $0.10/1M input, $0.20/1M output - requires allowing Meta to train on your traffic and is rate-capped at 60 requests/minute vs ~3,000 for standard. Weights: Meta announced on August 10, 2026, that Muse Spark 1.2 weights would be released openly “in the coming weeks.” Agents on Rails benchmark (Aug 2026, Le Mans round). 76.2% accuracy on 63 runs - eight points below GPT-5.6 Sol and about 3x Sol’s mean cost. The score is credible but not class-leading; watch for independent replication once the open weights and local tooling land.
Your workload
API cost for your workload
| Provider | Rate ($/1M) | Monthly cost |
|---|---|---|
| OpenRouter may be stale | $1.25 in · $4.25 out | $0.03 cheapest |
Monthly cost is an estimate from list prices and your workload - verify against the provider before committing. Cached fraction applies the cache rate to that share of input.
Can you run it locally?
Muse Spark 1.2 has no published quantization that fits a rig one person can buy, so there's no local-hardware recommendation and no break-even to compute. The honest answer is the per-provider API cost above.
Proprietary dense coding model from Meta, released August 5, 2026 alongside the Muse Code terminal agent. Native 1M-token context. Mandatory reasoning with five effort levels ( minimal, low, medium, high, xhigh; medium default). Multimodal: text, image, video, audio, and PDF input; text output. Pricing (Meta Model API standard): $1.25/1M input, $4.25/1M output; cached input $0.15/1M. Pricing (Contributor tier): $0.10/1M input, $0.20/1M output - requires allowing Meta to train on your traffic and is rate-capped at 60 requests/minute vs ~3,000 for standard. Weights: Meta announced on August 10, 2026, that Muse Spark 1.2 weights would be released openly “in the coming weeks.” Agents on Rails benchmark (Aug 2026, Le Mans round). 76.2% accuracy on 63 runs - eight points below GPT-5.6 Sol and about 3x Sol’s mean cost. The score is credible but not class-leading; watch for independent replication once the open weights and local tooling land.
See the model card for the full architecture notes and any cloud subscription plans.