Meta / meta/muse-glimmer-30b

Muse Glimmer 30B - access through LLMTR

Muse Glimmer 30B is Meta's 29.8-billion-parameter open-weight model, released under Apache 2.0. It is tuned for agents that keep running for hours: reliable tool use, state tracking across long tasks, and recovery after a failed step are its priorities. Its 131,072-token context window holds a large codebase or an entire tool loop in a single session. Alongside text it accepts image and video input; it makes tool/function calls and produces both JSON object output and schema-conforming structured output. It reasons step by step before answering, and you tune that effort with `reasoning_effort` across none, minimal, low, medium, high, xhigh, and max. Note that `none` does not turn reasoning off — it is simply the lowest setting; the model always produces a reasoning trace, and those tokens are billed as output. Repeating the same prompt triggers a prompt-cache discount.

Technical specifications

Canonical IDmeta/muse-glimmer-30b
ProviderMeta
Context window131,072 tokens
OperationsCHAT_COMPLETIONS
Modalitiestext, image, video

Pricing

An 8% platform margin applies to credit top-ups; model usage prices are not separately marked up.

OperationMetricUnitPrice
CHAT_COMPLETIONSINPUT_TEXTPER_1M_TOKENS$0.350000
CHAT_COMPLETIONSCACHE_READPER_1M_TOKENS$0.040000
CHAT_COMPLETIONSOUTPUT_TEXTPER_1M_TOKENS$1.50

Example usage

With existing OpenAI SDK flows, change only the base URL and model identifier.

curl https://llmtr.com/v1/chat/completions   -H "Authorization: Bearer llmtr-your_key"   -H "Content-Type: application/json"   -d '{"model":"meta/muse-glimmer-30b","messages":[{"role":"user","content":"Hello"}]}'

Related models