Muse Spark Open
Muse Spark Open is Meta's open-weight multimodal reasoning model, a sparse mixture-of-experts with 816 billion total and 42 billion active parameters.
input · usd / 1m tok
$1.25
output · usd / 1m tok
$4.25
context window
1.05mtokens
modalities · in → out
takes text and image in, gives text out
pricing
base api ratesusd per 1m tokens
- input
- $1.25
- output
- $4.25
- cache read
- $0.15
providers
Fireworks hosts Muse Spark Open. Impossibl routes your request intelligently and handles failover automatically when a provider is down.
- Fireworks
specifications
takes text and image in, gives text outmeta/muse-spark-open
- creatorMeta
- model typeText generation
- context window1,048,576 tokens
- maximum outputnot published
- input
takes text and image in, gives text out
- outputtext
- tool callingnot verified
- releasedoct 15, 2026
- endpoint
/v1/chat/completions
unknown means the public catalog does not establish that fact. a model's documented capabilities can differ from what a particular api supports. see the model documentation for integration details.
try it · your prompt, this model, live
# responses API
import os
from openai import OpenAI
client = OpenAI(
api_key=os.environ["IMPOSSIBL_API_KEY"],
base_url="https://api.impossibl.com/v1",
)
resp = client.responses.create(
model="meta/muse-spark-open",
input="",
)
print(resp.output_text)
edit the prompt, then run
benchmarks
intelligence index
38.9#30 of 110 models on the gateway
- omniscience24.6#12 of 86
- gpqa89.0%#32 of 82
- humanity’s last exam43.9%#13 of 89
- GPQA89.0%
- MMMU-Pro80.9%
- AA-LCR78.3%
- SciCode58.4%
- AutomationBench Partial49.7%
- Humanity’s Last Exam43.9%
- GDP.pdf All-pass18.6%
- CritPt14.9%
- Intelligence Index38.87
- Omniscience Index24.65
- GDPval Elo1401.12
- legal index (API)44.5
- economics index (API)51
- engineering index (API)40.4
- intelligence index (API)38.9
- strategy and ops index (API)42.7
- finance and accounting index (API)42.5
each configuration is listed separately; these evaluations describe the tested model configuration, and performance and costs can differ on impossibl. unmeasured scores are omitted and api values keep their source units. last checked 2026-10-19.
compare Muse Spark Open with other models
frontier labs and open weights, best-ranked firstusd / 1m tok · in / out
- anthropic/claude-opus-5-5$4.00 / $20.00compare →
- openai/gpt-6-astra$10.00 / $50.00compare →
- xiaomi/mimo-v2.6-pro$0.435 / $0.87compare →
- qwen/qwen3.8-max$2.00 / $6.00compare →
frequently asked questions
What is Muse Spark Open?
Muse Spark Open is Meta's open-weight multimodal reasoning model, a sparse mixture-of-experts with 816 billion total and 42 billion active parameters. It is the downloadable sibling of the hosted Muse Spark line, aimed at coding and long-horizon agentic work that you can run on your own hardware or any inference host.
How much does Muse Spark Open cost?
Muse Spark Open costs $1.25 per million input tokens and $4.25 per million output tokens through the Impossibl API. Cache reads cost $0.15 per million tokens. All rates are in USD.
What is the context length of Muse Spark Open?
Muse Spark Open has a 1,048,576-token context window through the Impossibl API.
What inputs and outputs does Muse Spark Open support?
Muse Spark Open accepts text and image as input. Muse Spark Open returns text.
How do I use Muse Spark Open through an API?
Muse Spark Open is available through the Impossibl API. Use meta/muse-spark-open as the model identifier with /v1/chat/completions. The quickstart explains how to create an API key and send your first request.
When was Muse Spark Open released?
Meta released Muse Spark Open on October 15, 2026.
this page describes Muse Spark Open as listed by the impossibl public api. prices and availability come from that catalog, refreshed here every five minutes. this is an impossibl offering, not a comparison of independent hosting-provider quotes. model descriptions and release dates, where available, use the creator sources linked under the answers above.
view the source catalog ↗model as markdown ↗report a correction →