Meta launches Muse Ember, a low-cost agent model priced at $0.10 per million input tokens
Meta has released Muse Ember, the smallest model in its Muse family, aimed at high-volume agent work such as tool calls, extraction and routing. It offers a 1M-token context and 512K output tokens, priced from $0.10 per million input tokens, and is available through Meta's API and Google and Oracle clouds.
Meta published Muse Ember on 9 November, describing it as its smallest and most efficient model. It is designed for the high-volume plumbing of agent systems: tool calls, coding subtasks, routing, extraction and parsing of long documents.
The model handles text, images, audio and video natively, can operate a computer from screenshots, and supports a one-million-token context window with up to 512,000 output tokens. Meta says it is the first model trained using Muse Wonder Full. The page does not mention open weights; access is via the Meta Model API as muse-ember-1.0, Muse Code, and Google and Oracle clouds.
Pricing is $0.10 per million input tokens and $0.50 per million output tokens for prompts up to 200K tokens, rising to $0.30 and $1.50 beyond that. Cached input is listed at $0.004 per million.
On Meta's own benchmarks, Ember scores 37.8 per cent on AutomationBench-AA against 13.9 per cent for Claude Haiku 5.5 and 20.2 per cent for GPT-6 Luna, and 40.9 per cent on Terminal-Bench 4.0. It trails Haiku on FrontierCode 2.0-AA, 27.2 against 34.9 per cent. Meta puts its cost per Artificial Analysis Index task at about $0.14, against $2.39 for Muse Spark 1.3.
Meta says Ember stayed below its framework's high-risk thresholds for chemical and biological, cyber and loss-of-control risks.
Why it matters
Small models decide the unit economics of agents. At these prices, running thousands of routine steps per workflow becomes affordable, and the competition among Meta, Anthropic and OpenAI in the cheap tier is intensifying.
Indicops view
For Indian teams building document-heavy workflows in banking, insurance and public services, a model in this price band is worth testing against current small-model choices. Validate on your own data, since vendor benchmarks vary by task, and check data-residency terms before production use.
More in Models
Back to the newsroomGoogle launches Gemini 4 Hydrogen, a $1.50-per-million-token model for coding and agents
Google has released Gemini 4 Hydrogen, a faster and cheaper tier of its Gemini 4 family aimed at everyday coding and agentic work. It costs $1.50 per million input tokens and $7.50 per million output tokens across a 1M-token window, and is rolling out through the Gemini API, AI Studio and consumer products.
Alibaba opens a two-week preview of Qwen4-Max with a 1M-token context, but no benchmarks or weights
Alibaba Cloud has opened a time-limited preview of Qwen4-Max, its next flagship model, through the Token Plan only. It offers a 1M-token context and up to 393,216 output tokens, but the documentation lists no pricing, benchmarks or downloadable weights, and the model retires on 25 November.
Anthropic releases Claude Fable 5.2 and Mythos 5.2, 30 per cent faster and 20 per cent cheaper
Anthropic has released Claude Fable 5.2 for general use and Mythos 5.2 for vetted cyber and life-sciences organisations. The company says output is up to 30 per cent faster than the 5.1 generation, at $8 per million input tokens and $40 per million output tokens, with lower rates of safeguard blocks.