Meta · in Borg
MetaMuse Spark 1.3
Muse Spark 1.3 is the newest version of Meta's Muse Spark reasoning line.
Reasoning1,048,576 tokenstext, image, video, audio, PDF
Included in Borg
Download Borg Borg for BusinessAt a glance
| Specification | Value |
|---|---|
| Provider | Meta |
| Family | Muse Spark |
| Release date | 2026-09-02 Vendor source |
| Context window | 1,048,576 tokens Vendor source |
| Input modalities | text, image, video, audio, PDF |
| Output modalities | text |
| Reasoning | Yes |
| Open weights | No |
| Borg lane ids | muse-spark-1.3, meta/muse-spark-1.3 |
| Versions in Borg | Muse Spark 1.1 · Muse Spark 1.2 · Muse Spark 1.3 |
| Median time to first token (gateway proxy) on Borg | 5938 ms |
| Median tokens per second (end to end) on Borg | 26.2 output tokens per second |
Not published by the vendor: thinking modes, knowledge cutoff, fusion roles.
What Muse Spark 1.3 is good for
Muse Spark 1.3 is the newest version of Meta's Muse Spark reasoning line. It follows the coding-centered 1.2 with stronger long-horizon agent behavior, cleaner coding output and better handling of interrupted or conflicting instructions. Meta says it uses fewer tool calls and tokens than 1.2 in its own engineering comparisons; unlike earlier versions, its standard tier also supports max reasoning effort for harder jobs. The model takes text, images, video, audio and PDFs and produces text, though Meta warns that audio understanding in this version is not yet fully supported. It is a practical choice for sustained coding agents, browser workflows and tasks that require keeping many requirements straight. Borg includes it in the subscription; Borg Fusion can route to it or you can choose its lane directly.
- long-horizon coding with changing requirements
- multi-step browser and tool workflows
- large-context instruction following
- reasoning-intensive repository tasks
How Borg uses it
Muse Spark 1.3 is available as a subscription catalog lane in Borg. The catalog does not publish which Fusion profiles route to this model, so no routing role is claimed here. You can pick it directly by setting the OpenAI-compatible gateway model field to a listed lane id. Check the current catalog for availability before building a workflow around a lane.
OpenAImodel: "muse-spark-1.3"Measured on Borg
| Metric | Median | 95th percentile | Samples |
|---|---|---|---|
| Time to first token (gateway proxy) | 5938 ms | 32423 ms | 1912 |
| Tokens per second (end to end) | 26.2 output tokens per second | 77.1 output tokens per second | 1928 |
Time to first token (gateway proxy)
Output tokens per second (whole turn)
- Time to first token (gateway proxy): time until the first streamed response event reaches the gateway; our closest available measure of time to first token
- Tokens per second (end to end): tokens per second, end to end: provider-reported output tokens divided by the whole turn's elapsed time, including queueing and reasoning; not a streaming per-token rate
Completion speed covers the whole turn including queueing and reasoning time; it is not a streaming tokens-per-second figure.
Measured on Borg: SearchAtlas AI Gateway production request logs. 30-day window ending 2026-10-08; 2340 requests. Minimum 100 samples per published request metric.
Benchmarks
| Benchmark | Vendor-reported score | Source |
|---|---|---|
| JobBench (max) | 64.9 | Vendor source |
| DeepSearchQA (max) | 90.3 | Vendor source |
| AutomationBench (max) | 49.6 | Vendor source |
| DeepSWE v1.1 (max) | 75.4 | Vendor source |
| Terminal-Bench 2.1 (max) | 88.8 | Vendor source |
Pricing
Included in Borg subscription. Vendor API list price: $1.25 input / $4.25 output per 1M tokens. Vendor price source.
Compare subscription and API pricing · Cost Savings Calculator
FAQ
What is Muse Spark 1.3?
Muse Spark 1.3 is listed in Borg's subscription catalog as muse-spark-1.3. Muse Spark 1.3 is the newest version of Meta's Muse Spark reasoning line.
Is Muse Spark 1.3 in Borg?
Yes. Muse Spark 1.3 is listed in Borg's subscription catalog under the lane muse-spark-1.3. You can choose the lane directly, or explore the available Fusion profiles. Availability and model details can change; check the current catalog before deployment.
What context window does Muse Spark 1.3 have?
The published context window is 1,048,576 tokens. Consult the cited sources for details and current limits. Vendor source
Is Muse Spark 1.3 good for coding?
Vendor-published evaluations appear in the benchmarks section; their settings may differ from your tasks. Its listed uses include long-horizon coding with changing requirements, multi-step browser and tool workflows, large-context instruction following, reasoning-intensive repository tasks. Try the lane for your own tasks, and compare the output with your requirements before choosing a default.
How fast is Muse Spark 1.3 on Borg?
Measured on Borg: SearchAtlas AI Gateway production request logs. 30-day window ending 2026-10-08; 2340 requests. Minimum 100 samples per published request metric.
How much does Muse Spark 1.3 cost on Borg vs the API?
This model is included in the Borg subscription. The vendor API list prices appear below with a source link. See the pricing explainer and cost savings calculator to compare billing approaches.
Sources
Related
Other versions of Muse Spark: Muse Spark 1.1 · Muse Spark 1.2
Other Meta models: Muse Spark 1.1 · Muse Spark 1.2
Models sharing a Fusion role: Not published