Muse Spark 1.2 API: Meta's flagship model on FastMetal
Meta's Muse Spark 1.2 is available on FastMetal as muse-spark-1.2. Until now the only Meta model we served was the small Muse Glimmer 30B; this is the flagship line.
What you get
- A top-ranked model. On August 29, 2026 it was #4 overall on the public arena text leaderboard (score 1498), the highest-ranked model we served that day. Rankings move as the board updates; the rankings by use case and the model page show where it stands now.
- A 1,048,576-token context. A long spec or a whole codebase fits in one request.
- One upstream endpoint: Meta's own. Requests are not spread across hosts that run the model at different quantizations, so output quality stays consistent from run to run.
Calling it
Set model to muse-spark-1.2. The base URL and API key are the same as for every other model on FastMetal:
curl https://api.fastmetal.ai/v1/chat/completions \
-H "Authorization: Bearer $FASTMETAL_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "muse-spark-1.2",
"messages": [{"role": "user", "content": "List three contradictions in this spec."}]
}'
With the OpenAI SDK, point base_url at https://api.fastmetal.ai/v1 and change model. Nothing else changes. To try it without code, pick it in chat.
Comparing it
Model comparison puts real answers from different models to the same prompts side by side. Two pairs worth a look:
- Muse Glimmer 30B vs Muse Spark 1.2, Meta's small model against its flagship.
- Muse Spark 1.2 vs Muse Spark 1.3. Meta has since released Muse Spark 1.3, and 1.2 stays in the catalog beside it.
Pricing
Billed per million input and output tokens, from a prepaid balance, with a $5 minimum top-up and no monthly fee. Current rates are on pricing and the model page.