Muse Ember

Muse Ember is Meta's smallest and most efficient reasoning model, built for complex reasoning execution and everyday agent work. It accepts text, images, video, audio, and PDF documents and returns text, with a 1M-token context window.

Input / output priceFree: $0.00 / $0.00 per 1M
Context: 1M tokens
Released: Nov 9, 2026

Try Respan for free

Agent setup

Paste into Claude Code, Cursor, or Codex

Help me set up the Respan gateway: fetch and follow https://www.respan.ai/docs/agent-setup.md. Use model "meta/muse-ember:free".

from openai import OpenAI

client = OpenAI(

    base_url="https://api.respan.ai/api/",

    api_key="YOUR_RESPAN_API_KEY",

)

response = client.chat.completions.create(

    model="meta/muse-ember:free",

    messages=[{"role": "user", "content": "Hello!"}],

)

print(response.choices[0].message.content)

import OpenAI from "openai";

const client = new OpenAI({

  baseURL: "https://api.respan.ai/api/",

  apiKey: "YOUR_RESPAN_API_KEY",

});

const response = await client.chat.completions.create({

  model: "meta/muse-ember:free",

  messages: [{ role: "user", content: "Hello!" }],

});

console.log(response.choices[0].message.content);

curl https://api.respan.ai/api/chat/completions \

  -H "Authorization: Bearer YOUR_RESPAN_API_KEY" \

  -H "Content-Type: application/json" \

  -d '{

    "model": "meta/muse-ember:free",

    "messages": [{"role": "user", "content": "Hello!"}]

}'

Read docs

Providers

2 routes from 1 provider, cheapest first, priced per 1M tokens after gateway discounts.

Provider	Context	Input	Output	Cache read	Capabilities
MetaFree`meta/muse-ember:free`	1M	$0.00/M	$0.00/M	$0.00/M
Meta`meta/muse-ember`	1M	$0.10/M	$0.50/M	$0.004/M

Pricing

Averaged across 2 routes from 1 provider, in USD per 1M tokens after gateway discounts.

Average input: $0.05 per 1M
Average output: $0.25 per 1M
Average cache read: $0.0020 per 1M

Cost by volume

Total cost from 0 to 1T tokens on the cheapest and most expensive provider.

Highest$0.10/MMetaLowest$0.00/MMeta

Throughput and latency

30-day averages from requests through the Respan gateway. Under each figure, every model on the gateway by its average, this one in blue.

Output speed

Higher is better

2495tokens/sacross providers

Faster than 98% of the other 430 models on the gateway

Time to first token

Lower is better

2.5sacross providers

Faster than 32% of the other 119 models on the gateway

Response latency

Lower is better

3.1sacross providers

Faster than 45% of the other 492 models on the gateway

Benchmarks

The Respan Index plus 3 published scores from 3 sources, each linked to where it was published.

61.9/ 100

Respan Index#15 of 169

169 ranked models. The shaded band is this model's 95% interval, 38.6 to 79.1. GPT-6 Sol and GPT-5.6 Sol fall inside it, so the gap is within the margin of error.

Published scores

Benchmark	Score	Source	Published
Terminal-Bench 4.0Vals AI	39.4%	Vals AI	Nov 10, 2026

Benchmark	Score	Source	Published
SimpleBenchSimpleBench	72.7%	SimpleBench	Nov 10, 2026

Benchmark	Score	Source	Published
LMArena WebDevLMArena	1609 Elo · #29 of 118	LMArena	Nov 10, 2026

How the Respan Index works

Apps

Use Muse Ember in the tools you already run. Each guide points the app at the Respan gateway.

Setup guides for more apps are in the integration docs.

Muse Ember FAQ

Short answers, built from the data on this page.

Muse Ember is a model from Meta, released on Nov 9, 2026. Muse Ember is Meta's smallest and most efficient reasoning model, built for complex reasoning execution and everyday agent work. It accepts text, images, video, audio, and PDF documents and returns text, with a 1M-token context window.

On the Respan gateway, Muse Ember costs from $0.00 per 1M input tokens and $0.00 per 1M output tokens on Meta, after gateway discounts. Averaged across its 2 routes from 1 provider, it costs $0.05 per 1M input tokens and $0.25 per 1M output tokens.

Muse Ember is served by Meta through 2 routes.

Muse Ember accepts up to 1M tokens of context on its largest provider route.

Point any OpenAI-compatible SDK at https://api.respan.ai/api/ with your Respan API key and set the model to "meta/muse-ember:free". To set it up with a coding agent, copy the agent setup prompt at the top of this page.

Muse Ember scores 61.9 out of 100 on the Respan Index v1.1, ranking #15 of 169 models. By category: Coding 51.0, Agents 73.0 (estimated), Reasoning 63.9, Knowledge & language 63.7 (estimated), and Human preference 56.7. The Respan Index combines independent and lab-reported benchmark results into one score, using benchmark data collected by Respan.

Muse Ember has 3 published benchmark scores, including Terminal-Bench 4.0 39.4%, SimpleBench 72.7%, and LMArena WebDev 1609. Each score links to where it was published; the data is collected by Respan.