Muse Spark 1.2
Muse Spark 1.2 is a chat-oriented large language model listed under Meta in our catalog; published specifications for this release are limited.
API Pricing
| Provider | Input / 1M | Output / 1M | Cached / 1M |
|---|---|---|---|
| $1.25 | $4.25 | $1.00 | |
| $1.25 | $4.25 | $0.150 |
Prices updated daily. Last check: Oct 9, 2026
Muse Spark 1.2 pricing by provider
Input, output, and batch rates, plus alternatives, for one provider at a time.
Performance & Benchmarks
Source: Artificial Analysis →Reasoning & Knowledge
- GPQA Diamond90.4%
- Humanity's Last Exam45.5%
Coding
- SciCode57.4%
Agentic & Tool Use
- Terminal-Bench v2.180.1%
- τ-bench Banking34.8%
Instruction & Long Context
- Long-Context Reasoning79.0%
Benchmarks measured Oct 2026. Scores are independent evaluations, not vendor-reported.
Model Details
General
- Creator
- Meta
- Modalities
- Text
Capabilities
- Open Source
- No
Strengths & Limitations
Strengths
- Chat-tuned model, suitable for instruction-following and assistant-style interactions without additional fine-tuning
- Point revision (1.2) within the Muse Spark line, implying refinements over an earlier release in the same series
- Listed with per-provider pricing on this page, so cost can be compared across hosts side by side
- Attributed to Meta, whose chat models are commonly available through multiple third-party inference hosts
- Straightforward text-in, text-out interface that fits standard chat completion API integrations
Limitations
- We do not have a confirmed context window for this model, so long-document workloads need verification against provider documentation
- No verified benchmark scores are tracked for Muse Spark 1.2, making capability comparisons against peers difficult
- Throughput and time-to-first-token are recorded as zero in our data, meaning no latency measurement is available
- Multimodal support, tool calling, and structured output behavior are not documented in our metadata and must be confirmed with the provider
- Limited public documentation overall compared with more widely covered Meta model lines
Key Features
About Muse Spark 1.2
Common Use Cases
Muse Spark 1.2 is positioned as a general chat model, so the natural fit is conversational and text-generation work: assistant interfaces, drafting and rewriting, summarization of moderate-length inputs, question answering over provided context, and classification or extraction tasks expressed as prompts. Because we have no confirmed context window, treat long-context jobs such as full-codebase review or lengthy document analysis as unverified until you check the provider's stated limit. Similarly, agentic pipelines that depend on function calling or strict JSON output should be validated against provider documentation before you build on them. The most reliable way to scope this model is an A/B evaluation against a better-documented alternative on your own prompt set, comparing output quality alongside the provider costs listed above.
Frequently Asked Questions
How much does Muse Spark 1.2 cost to run?
Pricing varies by provider and by pricing type — input versus output tokens, batch versus real-time, and any cached-input discounts a host may offer. Rates also change frequently. Check the pricing table on this page for current per-provider figures rather than relying on a fixed number.
What is Muse Spark 1.2 best used for?
It is a chat model, so it suits assistant-style conversation, drafting and rewriting, summarization, question answering over supplied context, and prompt-based extraction or classification. For workloads that depend on a specific context length, image input, or tool calling, confirm support with your chosen provider first, since our metadata does not document those details.
What context window does Muse Spark 1.2 support?
We do not have a confirmed context window for this model in our database. Providers hosting it publish their own maximum context and output token limits, and those can differ between hosts, so check the provider's model documentation before designing around a specific token budget.
Why are the speed numbers listed as zero?
The throughput and time-to-first-token values in our entry come from Artificial Analysis and are recorded as zero, which indicates no measurement has been captured for Muse Spark 1.2 — not that the model is slow. Real-world latency also depends heavily on which provider serves the model and at what load.
How does Muse Spark 1.2 compare with other Meta chat models?
We do not track verified benchmark scores for Muse Spark 1.2, so a spec-based comparison is not possible from our data. The practical approach is to run it against a better-documented Meta chat model on a representative sample of your own prompts and weigh the quality difference against the per-provider costs shown above.