Meta model product
Muse Spark 1.1
1 is Meta Superintelligence Labs' multimodal reasoning model for agentic tasks.
Updated Aug 12, 2026. Default version: Muse Spark 1.1
LLMBoard score88.3Muse Spark 1.1
Coverage40%10 benchmark families
Context window1MTokens
Official input price$1.25Meta API
Muse Spark 1.1 Specifications
Technical details for the model's default version.
- Version
- Muse Spark 1.1
- Released
- Jul 9, 2026
- Knowledge cutoff
- Unknown
- Parameters
- N/A
- Context window
- 1M
- Max output
- 32K
- Inputs
- image, pdf, text, video
- Outputs
- text
- Open weights
- No
- License
- Proprietary
Muse Spark 1.1 Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Muse Spark 1.1 Benchmark Results
Benchmark scores for Muse Spark 1.1.
| Job Bench | 54.7% | 01 | 2 | 100.0% | C | |
| MCP Atlas | 88.1% | 01 | 31 | 100.0% | C | |
| Toolathlon | 75.6% | 01 | 31 | 100.0% | C | |
| BabyVision | 76.3% | 02 | 9 | 87.5% | C | |
| Finance Agent v2 | 57.2% | 02 | 26 | 96.0% | C | |
| Humanity's Last Exam | 62.1% | 04 | 93 | 96.7% | C | |
| CharXiv-R | 88.4% | 06 | 48 | 89.4% | C | |
| OSWorld-Verified | 80.8% | 06 | 23 | 77.3% | C | |
| Terminal-Bench 2.1 | 80.0% | 11 | 19 | 44.4% | C | |
| SWE-Bench Pro | 61.5% | 12 | 45 | 75.0% | C | |
| DeepSWE 1.1 | 53.0% | 13 | 20 | 36.8% | B |
Muse Spark 1.1 Arena Results
Preference and agent-evaluation results for the default version.
| document style control | overall | 08 | 1482.8 | 2,054 | N/A | |
| text style control | overall | 10 | 1488.7 | 16,226 | N/A | |
| agent task outcome explicit | overall | 11 | 0.1 | N/A | 37.9K | |
| document | overall | 13 | 1471.7 | 2,054 | N/A | |
| text | overall | 14 | 1476.7 | 16,226 | N/A | |
| vision | overall | 16 | 1295.5 | 4,661 | N/A | |
| agent tool hallucination | overall | 17 | 0.0 | N/A | 2M | |
| webdev | overall | 17 | 1537.5 | 5,241 | N/A | |
| vision style control | overall | 19 | 1282.2 | 4,661 | N/A | |
| agent bash recovery steps | overall | 21 | 0.1 | N/A | 39.1K | |
| agent | overall | 22 | 0.0 | N/A | 2.1M | |
| text factuality | overall | 27 | 1465.3 | 16,201 | N/A | |
| agent praise complaint | overall | 30 | -0.0 | N/A | 14.6K | |
| agent steerability | overall | 31 | -0.0 | N/A | 55.6K |
Muse Spark 1.1 Runtime Performance
Provider-specific output speed and catalog latency for Muse Spark 1.1. Runtime does not affect the capability score.
| Meta Model API | 13.687 tok/s | 29.373 s | 1M | 131.1K |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Muse Spark 1.1 Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- $1.25 input, $4.25 output per 1M
- Official provider
- Meta
- Lowest third-party
- From $1.25 input, $4.25 output per 1M via NanoGPT
- Tracked offerings
- 10
| NanoGPT | meta/muse-spark-1.1 | global | $1.25 | $4.25 | 1M | |
| LLM Gateway | muse-spark-1.1 | global | $1.25 | $4.25 | 1M | |
| Pioneer | meta/muse-spark-1.1 | global | $1.25 | $4.25 | 1M | |
| Abacus | muse-spark-1.1 | global | $1.25 | $4.25 | 1M | |
| EmpirioLabs AI | muse-spark-1-1 | global | $1.25 | $4.25 | 1M | |
| Meta | muse-spark-1.1 | global | $1.25 | $4.25 | 1M | |
| Kilo Gateway | meta/muse-spark-1.1 | global | $1.25 | $4.25 | 1M | |
| Merge Gateway | meta/muse-spark-1.1 | global | $1.25 | $4.25 | 1M | |
| Vercel AI Gateway | meta/muse-spark-1.1 | global | $1.25 | $4.25 | 1M | |
| OpenRouter | meta/muse-spark-1.1 | global | $1.25 | $4.25 | 1M |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Muse Spark 1.1 Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
| Muse Spark 1.1 | 88.3 | N/A | 1M | 32K | No | Proprietary |
What is Muse Spark 1.1?
Key information about Muse Spark 1.1 and its available data.
1 is Meta Superintelligence Labs' multimodal reasoning model for agentic tasks. It improves tool and computer use, coding, and multimodal understanding over Muse Spark, can orchestrate parallel subagents, and actively manages a 1,048,576-token context window.
The Meta Model API supports text, image, PDF, and video inputs, up to 131,072 output tokens, structured output, parallel tool calling, and built-in web search with citations.
Data as of 2026-08-11.
Muse Spark 1.1 vs nearby models
Open a comparison with the three ranked models immediately above and below this model.
Models similar to Muse Spark 1.1
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
ME
Muse Spark 1.2
Meta
ME
Muse Spark
Meta
ME
Muse Glimmer 30B
Meta
ME
Llama 4 Maverick
Meta
ME
Llama 3.1 405B
Meta
ME
Llama 3.3 70B
Meta
FAQ
Common questions about Muse Spark 1.1.
When was Muse Spark 1.1 released?
Muse Spark 1.1's default version was released on Jul 9, 2026.
How much does Muse Spark 1.1 cost?
Muse Spark 1.1's official API price is $1.25 per million input tokens and $4.25 per million output tokens via Meta. The lowest tracked third-party offer starts at $1.25 input and $4.25 output via NanoGPT.
Who created Muse Spark 1.1?
Muse Spark 1.1 was created by Meta.
What is the context window for Muse Spark 1.1?
The default version has a 1M token context window.
Is Muse Spark 1.1 open weight?
No. The default version is not marked as having publicly available weights.
How many API providers offer Muse Spark 1.1?
10 provider offerings are linked to the default version.
What models should I compare Muse Spark 1.1 with?
Nearby ranked alternatives include Claude Mythos, Kimi K3, Qwen3.8 Max.