Meta: Muse Spark Open

{"about":"### Meta 用於開發與長任務的模型\n\nMuse Spark Open 是公開權重的 Muse Spark 版本，著重程式設計與代理工作流程。Meta 強調需要跨多個步驟保持目標的任務：了解專案、規劃修改、準備程式碼，並讓結果符合原始需求。在 Ailoot，模型接受文字訊息並傳回文字。\n\n根據 Meta，Muse Spark Open 總計含 8160 億參數，每個 token 啟用 420 億參數。這些數值描述模型的整體規模與處理 token 時使用的部分。公開權重本身不代表使用者必須下載權重或自行安排本機執行，才能使用 Ailoot。\n\n### 程式碼與需求工作\n\n文字請求可包含任務描述、原始碼片段、環境資訊與預期行為。分析錯誤時，適合附上實際訊息、最小範例與修改限制。模型可準備說明、修正方案與驗證建議；執行程式碼及確認結果仍是工作流程中的另外步驟。\n\n設計修改時，可規定結果順序：先指出受影響部分，再描述方案，最後提供程式碼。進行審查時，請將需求與修改片段一併附上，使討論根據系統行為，而不只是撰寫風格。這些用法符合模型所宣稱的程式設計與多步驟解題方向。\n\n### Ailoot 的上下文\n\nMuse Spark Open 標示上下文為 1,048,576 個 token，可放入大量文字資料及任務歷史。大型視窗適合一併檢視需求、文件與數個相關模組；仍應明確保留檔名及片段邊界，以便答案引用。\n\nMuse Spark Open 基礎模型具有多模態能力，但這張模型卡目前的 Ailoot 方法只處理文字。本說明並未宣稱能分析附加影像、音訊或影片，也未宣稱能控制程式。Meta 的代理示範是在另有工具的獨立環境中運作；文字答案本身不代表應用程式或儲存庫內的修改已完成。","alias_of":null,"aliases":[],"author":"Meta","canonical_model":"meta/muse-spark-open","ctx":null,"formats":["chat","gemini","messages","responses"],"id":"meta/muse-spark-open","input":["text"],"kind":"text","max_out":null,"methods":[{"cache_read":0.3,"cache_write":null,"cache_write_1h":null,"cache_write_5m":null,"conditional_meter":null,"ctx":null,"family":"anthropic","fast_multiplier":1,"flex_multiplier":1,"fmt":"messages","http":"POST","img_in":null,"img_out":null,"in":2.5,"input":["text"],"label":"Anthropic Messages","max_out":null,"max_tool_calls":null,"out":8.5,"output":["text"],"path":"/v1/messages","request_model_id":"meta/muse-spark-open","route_label":"Ultra","status_reads":[],"tiers":[],"tool_prices":null,"unit":"token","units":[]},{"cache_read":0.3,"cache_write":null,"cache_write_1h":null,"cache_write_5m":null,"conditional_meter":null,"ctx":null,"family":"openai","fast_multiplier":1,"flex_multiplier":1,"fmt":"responses","http":"POST","img_in":null,"img_out":null,"in":2.5,"input":["text"],"label":"OpenAI Responses","max_out":null,"max_tool_calls":null,"out":8.5,"output":["text"],"path":"/v1/responses","request_model_id":"meta/muse-spark-open","route_label":"Ultra","status_reads":[],"tiers":[],"tool_prices":null,"unit":"token","units":[]},{"cache_read":0.3,"cache_write":null,"cache_write_1h":null,"cache_write_5m":null,"conditional_meter":null,"ctx":null,"family":"openai","fast_multiplier":1,"flex_multiplier":1,"fmt":"chat","http":"POST","img_in":null,"img_out":null,"in":2.5,"input":["text"],"label":"OpenAI Chat","max_out":null,"max_tool_calls":null,"out":8.5,"output":["text"],"path":"/v1/chat/completions","request_model_id":"meta/muse-spark-open","route_label":"Ultra","status_reads":[],"tiers":[],"tool_prices":null,"unit":"token","units":[]},{"cache_read":0.3,"cache_write":null,"cache_write_1h":null,"cache_write_5m":null,"conditional_meter":null,"ctx":null,"family":"gemini","fast_multiplier":1,"flex_multiplier":1,"fmt":"gemini","http":"POST","img_in":null,"img_out":null,"in":2.5,"input":["text"],"label":"Gemini generateContent","max_out":null,"max_tool_calls":null,"out":8.5,"output":["text"],"path":"/v1beta/models/meta/muse-spark-open:generateContent","request_model_id":"meta/muse-spark-open","route_label":"Ultra","status_reads":[],"tiers":[],"tool_prices":null,"unit":"token","units":[]}],"model_aliases":[{"canonical_model_id":"meta/muse-spark-open","request_model_id":"meta/muse-spark-open","route_label":"Ultra"}],"name":"Meta: Muse Spark Open","price":{"in":2.5,"in_label":null,"out":8.5,"out_label":null,"sort_in":2.5,"sort_out":8.5,"varies":false},"public_id":"meta/muse-spark-open","short":"Meta 公開權重的模型，用於程式設計、多步驟推理與大量文字上下文處理。"}