StepFun Releases Step 5 Preview Open Weights: 600B MoE, 27B Active, 1M Context
https://www.marktechpost.com/2026/10/16/stepfun-releases-step-5-preview-open-weights-600b-moe-27b-active-1m-context/📌 StepFun 開權重 Step 5 Preview：600B MoE、僅啟動 27B、上下文上看 1M
TL;DR：StepFun 開源 600B 參數的 agentic MoE 模型，僅啟動 27B，提供可自架的長任務 coding/金融選項。
一個總參數 600B、每次推理卻只喚醒 27B 的模型，context 長度上看 1M tokens，而且權重直接開放下載——這是 StepFun 為「長時間 agentic 任務」量身打造的旗艦模型。
🤔 為軟體工程、知識工作與金融而生
Step 5 Preview 是 StepFun 針對 agentic 工作推出的旗艦模型，定位在軟體工程、專業知識工作與金融三大場景。StepFun 表示這個模型能在大型程式碼庫中導航、協調跨多檔案的修改、執行程式碼並根據結果持續迭代，也能分析長文件、圖表、截圖與影片，產出結構化報告。API 於 9 月 20 日上線，權重則在 10 月 14 日跟進釋出，FP8 與 BF16 兩種 checkpoint 都已經放上 Hugging Face 與 ModelScope。
🧩 600B 總參數，27B 稀疏啟動
Step 5 Preview 是一個稀疏 Mixture-of-Experts 模型，總參數量 600B，每個 token 只啟動 27B，context 長度最高可達 1M tokens。部署面，vLLM 與 SGLang 都提供了附帶 OpenAI 相容 API 的專用 Docker image；在 Blackwell 架構上，expert-parallel 模式會使用 MegaMoE backend。StepFun 建議的推理參數是 temperature 1.0、top_p 0.95、top_k 0。
📊 跑分細節：自評用 High reasoning，對手用 Max effort
需要留意的是，StepFun 自己的評測是在 High reasoning 設定下進行,而圖表裡的對手模型是在 Max effort 下測試,嚴格說起來並非同條件比較。以下數字出自 Hugging Face 模型卡,其中 Intelligence Index、AA-LCR、Humanity’s Last Exam、GDPval-AA 是來自 Artificial Analysis 透過 OpenRouter 取得的獨立測試結果，SRE Bench 與 Harvey’s Legal Agent Benchmark 來自 Vals AI，FrontierFinance 則是 StepFun 自家對比圖表中的數字（由 CellCog 與 eesel 轉錄）：
| Benchmark | 來源 | Step 5 Preview |
|---|---|---|
| Intelligence Index | Artificial Analysis | 43.7 |
| AA-LCR | Artificial Analysis | 88.3% |
| Humanity’s Last Exam | Artificial Analysis | 46.5% |
| GDPval-AA | Artificial Analysis | 54.3% |
| SRE Bench | Vals AI | 0.4% |
| Harvey Legal Agent Benchmark | Vals AI | 9.2% |
| FrontierFinance | StepFun 自評 | 66.4% |
在 FrontierFinance 上，Step 5 Preview 的 66.4% 勝過 GLM-5.3 與 Kimi K3，但落後 Claude Opus 5 的 69.7%。而 SRE Bench 僅 0.4%、Harvey’s Legal Agent Benchmark 僅 9.2%，顯示在 SRE 維運與法律 Agent 這類任務上，這個模型的弱點相當明顯。
⚠️ 權重龐大，部署門檻高
FP8 checkpoint 單檔就有 650.32 GB，這是一個徹頭徹尾的資料中心級模型，自架需要對應規模的 GPU 叢集，並不適合個人或小團隊在單機上跑完整版本。
🎯 API 定價與整合方式
StepFun API 的計價是每 1M input tokens 1.00 美元、每 1M output tokens 2.70 美元，輸出計價含推理 token，cache 命中則降至每 1M 0.10 美元。API 支援 low、medium、high 三種 reasoning effort、工具呼叫（tool calling）與 JSON Schema 輸出，對已經在用其他 OpenAI 相容 API 的 agent 框架來說，切換成本不高，是值得納入 benchmark 清單的新選項。
🔗 來源
- 標題：StepFun Releases Step 5 Preview Open Weights: 600B MoE, 27B Active, 1M Context
- 作者／機構：Michal Sutter, MarkTechPost
- 連結：https://www.marktechpost.com/2026/10/16/stepfun-releases-step-5-preview-open-weights-600b-moe-27b-active-1m-context/
#StepFun #OpenWeights #MixtureOfExperts #LLM #AgenticAI #LongContext #CodingAgent #OpenSourceAI #HuggingFace #MoE
原始資料 MarkTechPost · 收集於 2026-10-17
摘要原文
StepFun has released Step 5 Preview as open weights. FP8 and BF16 checkpoints are live on Hugging Face and ModelScope . Step 5 Preview is a sparse Mixture-of-Experts model from StepFun built for long-horizon agentic work. It has 600B total parameters, 27B active per token and up to 1M tokens of context. For developers, this is a new self-hostable option for coding agents, document analysis and finance workflows. Step 5 Preview is StepFun’s flagship model for agentic work . StepFun positions it for software engineering, professional knowledge work and finance. The model navigates large codebases, coordinates multi-file changes, runs code and iterates on results. It also analyzes long documents, charts, screenshots and video to produce structured reports. The API launched September 20 . Weights followed on October 14 . Only 27B of the 600B parameters activate per token. StepFun’s tech team adds the internals: StepFun ran its own evaluations at the High reasoning setting. Rivals in its chart ran at Max effort , so the comparison is not like-for-like. Results reported on the Hugging Face card : Independent numbers come from Artificial Analysis via OpenRouter . The model scores 43.7 on the Intelligence Index. It hits 88.3% on AA-LCR and 46.5% on Humanity’s Last Exam. GDPval-AA lands at 54.3%. Vals AI results show clear weak spots. SRE Bench is 0.4%. Harvey’s Legal Agent Benchmark is 9.2%. On finance, StepFun’s chart shows 66.4% on FrontierFinance . That beats GLM-5.3 and Kimi K3, but trails Claude Opus 5 at 69.7%. Benchmark rows come from StepFun’s comparison chart, as transcribed by CellCog and eesel . This is a data-center model. The FP8 checkpoint alone is 650.32 GB . Listed configurations : vLLM and SGLang ship dedicated Docker images with OpenAI-compatible APIs. On Blackwell, expert-parallel modes use the MegaMoE backend. StepFun recommends temperature 1.0, top_p 0.95 and top_k 0. The StepFun API charges $1.00 per 1M input tokens and $2.70 per 1M output tokens. Cache hits cost $0.10 per 1M. Output pricing includes reasoning tokens. The API supports low, medium and high reasoning effort, tool calling and JSON Schema output. Check out the model weights on Hugging Face , GitHub repo , tech blog , BF16 weights , ModelScope and API docs . All credit goes to the researcher of this project. Also, feel free to follow us on Twitter and don’t forget to join our 150k+ML SubReddit and Subscribe to our Newsletter . Wait! are you on telegram? now you can join us on telegram as well. The post StepFun Releases Step 5 Preview Open Weights: 600B MoE, 27B Active, 1M Context appeared first on MarkTechPost .
由 claude-code/sonnet 自動生成