Meta Releases Muse Spark 1.3 for Text and Image Inputs
Meta announced Muse Spark 1.3 on September 2, 2026. This model processes text and image inputs to generate text outputs without confirmed open weights.
Release Details and Input Modalities
Meta announced the release of Muse Spark 1.3 on September 2, 2026. This date marks the official publication of the model update. The model is designed to accept both text and image inputs. Its primary function is to process these multimodal inputs to generate text outputs. The source confirms that the model handles these specific input types but does not provide details on context window size or maximum output token limits.
Weight Availability and Access Constraints
The release documentation states that Muse Spark 1.3 does not include open weights. This fact records that open weights are not confirmed as part of this specific release package. It does not establish that local installation is impossible, nor does it confirm that API access is mandatory. The source does not describe deployment architecture, hosting arrangements, or integration behavior. Pricing information is also absent from the provided facts. Therefore, the availability status remains limited to the confirmation that open weights were not included in this announcement.
Practical Evaluation Procedure
Engineers can evaluate Muse Spark 1.3 by designing a controlled test suite. First, prepare a set of standardized text prompts and corresponding image files. Second, submit these inputs to the model interface. Third, measure the generated text outputs for relevance and format compliance. Fourth, compare the results against predefined quality metrics. Finally, validate whether the model consistently processes both modalities as described. This procedure allows for an objective assessment of the model's stated capabilities without inferring unverified benefits.
- Test: Submit mixed text and image inputs to verify acceptance.
- Measure: Record response time and output length for each trial.
- Compare: Analyze output accuracy against reference standards.
- Validate: Confirm that text outputs align with input intent.
Primary sources
Frequently asked questions
Does Muse Spark 1.3 support open weights?
No, the release documentation explicitly states that the release does not include open weights. This indicates that open weights are not confirmed for this version.
What inputs can Muse Spark 1.3 process?
Muse Spark 1.3 accepts text and image inputs. It produces text outputs based on these inputs. The source does not specify other supported modalities.