meta logo
meta

Meta: Muse Glimmer 30B (batch)

来源介绍(英文)

Muse Glimmer 30B is a dense, open-weight multimodal model from Meta Superintelligence Labs, distilled from Muse Spark and optimized for autonomous agents on consumer hardware. It is suited for long-horizon...

模型概览

模型规格

上下文
131,072 tokens
最大输出
16,384 tokens
架构
text+image->text
分词器
Other
知识截止时间
暂未提供
内容审核
否
OPENROUTER

完整价格

同步自 OpenRouter。Token 价格均按每百万 Token 展示。

输入
$0.24
每百万 Token
输出
$0.96
每百万 Token
缓存读取
$0.032
每百万 Token
API

模型配置

默认参数

temperature
1
top_p
0.95
top_k
64

推理能力

是否必须推理
是
默认参数
medium
能力与模态
xhigh, high, medium, low
API

快速调用

在本地设置 OPENROUTER_API_KEY。Python 需安装 requests;JavaScript 在 Node.js 中运行。密钥应保存在服务端。

此请求使用基础模型 ID 提交异步批处理。返回任务 ID 表示已提交,不表示已完成。

API 文档

curl --fail-with-body https://openrouter.ai/api/beta/batches \
  -H "Authorization: Bearer $OPENROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"endpoint":"/v1/chat/completions","model":"meta/muse-glimmer-30b","requests":[{"custom_id":"request-1","body":{"messages":[{"role":"user","content":"Hello!"}]}}]}'

# Replace JOB_ID with the returned id to check progress.
curl --fail-with-body "https://openrouter.ai/api/beta/batches/JOB_ID" \
  -H "Authorization: Bearer $OPENROUTER_API_KEY"
能力与模态

输入 → 输出

输入
textimage
输出
text
推理能力

否 · xhigh · high · medium · low

API

支持的 API 参数

frequency_penaltyinclude_reasoninglogit_biasmax_tokensmin_ppresence_penaltyreasoningreasoning_effortrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_p
可用提供方

1 提供方

核验日期: 2026年11月7日

在 OpenRouter 查看实时 Provider

Provider 可用性、延迟、吞吐量和路由会持续变化,请前往来源页面查看实时运行数据。

OpenRouter

DeepInfra

bf16
上下文
131K
最大输出
16K
输入
$0.24
输出
$0.96
缓存读取
$0.032
缓存写入
暂未提供
meta

4 个模型

全部模型
meta logometa

Meta: Muse Ember

Muse Ember is the smallest and most efficient model from Meta that’s designed for complex reasoning execution and the work agents repeat all day: tool calls, coding agent subtasks, routing,...

上下文
1.0M
输入
text · image · video · file · audio
输出
text
输入: $0.1输出: $0.5每百万 Token
查看模型详情 →
meta logometa

Meta: Muse Spark Open

Muse Spark Open is a coding-optimized open-weight model. Trained specifically for agentic coding workflows, it delivers higher first-attempt accuracy, reliable tool calling, and deep codebase understanding across the repos developers...

上下文
1.0M
输入
text · image · video · file · audio
输出
text
输入: $1.25输出: $4.25每百万 Token
查看模型详情 →
meta logometa

Meta: Muse Wonder

Muse Wonder is a frontier reasoning model from Meta, designed for complex personal and professional agentic tasks. It accepts text, images, video, audio, and PDF documents and returns text, with...

上下文
1.0M
输入
text · image · video · file · audio
输出
text
输入: $4输出: $20每百万 Token
查看模型详情 →
meta logometa

Meta: Muse Voice Transcribe 1.0

Muse Voice Transcribe 1.0 is a synchronous speech-to-text model from Meta. It is suited for push-to-talk, endpointing, and speaker-aware transcription, with keyword biasing for domain terms and language biasing through...

上下文
暂未提供
输入
audio
输出
transcription
音频时长: $0.00005 每秒
查看模型详情 →