MiniMax-H3-Image-PE：MiniMax 為其圖像模型推出 27B 提示詞擴充器，卻未告知任何人

There is no launch post for MiniMax-H3-Image-PE. There is no entry on M​iniMax's news feed, no blog, no changelog line, no social post — on 2026-11-11 the company's news index lists H3, M3.1 Flash, Speech 2.8 and Music 3.0, and nothing about images at all. What exists instead is a Hugging Face repository whose first commit is timestamped 2026-11-10 at 14:38 UTC, a model card that never states a release date, and a license file dated November 10, 2026. That is the whole announcement. MiniMax-H3-Image-PE is a prompt-expansion model — a caption writer — built to feed its sibling MiniMax-H3-Image, the open-weights 33.1B image generator M​iniMax published on 2026-11-03. The image model got eight days of quiet; its prompt expander has had one. Everything below that carries a number comes from those two repositories, the license text, or M​iniMax's own API documentation, and where something is not knowable from those sources it is marked as such rather than guessed. MiniMax-H3-Image and MiniMax-H3-Image-PE are not on OrcaRouter as of 2026-11-11, so this is a self-hosted evaluation. the catalogue The rest of the catalogue is one key away when you want to compare the expansion against a text model you already trust.

The short version, for anyone deciding whether to care: this model does not draw anything. You give it a sentence, it gives you back a long English description of the image you meant plus a recommended aspect ratio, wrapped in a two-key JSON object. A 27B model exists to write that paragraph because MiniMax-H3-Image is unusually fussy about its input — the parent card says it "expects a detailed English description of the whole image" and truncates anything past 4096 tokens. If you have ever typed six words into an image model and gotten six words' worth of thinking back, that gap is what this repository is trying to close, and on 2026-11-10 M​iniMax open-sourced its side of the fix.

儲存庫裡實際上有什麼，以及時間戳記顯示了什麼

The repository is MiniMaxAI/MiniMax-H3-Image-PE on Hugging Face. Its creation timestamp is 2026-11-08 15:35 UTC; its first commit — unhelpfully titled "Init MiniMaxAI/MiniMax-H3-Image-PE" — landed two days later, on 2026-11-10 at 14:38 UTC, authored by M​iniMax staff accounts ryanlee-dev and 500kg. A second commit followed 38 minutes later adding a Diffusers usage section, and that one is a pull request merged from Hugging Face's own multimodalart, which tells you the diffusers integration was contributed from outside M​iniMax rather than shipped by it. Three pull requests were opened against the repo on its first day, all closed or merged. None of them is an announcement.

What is inside is a checkpoint in four safetensors shards totalling roughly 17.9 GB, a tokenizer, a chat template, a quantisation config, and — the part that matters more than the weights — a prompt/ directory containing the actual system prompt, task prompt and request parameters M​iniMax uses. There is also a single client script, scripts/expand_prompt.py. The repository declares 27.4B parameters in total across two tensor types: about 25.6B stored as U8 and 1.76B left in BF16. That split is explained by the quantisation config, which records NVFP4 with a group size of 16 and lists the modules held out of quantisation — the embedding table and the convolution and gating projections inside each linear-attention block.

The base model is stated plainly on the card and in the metadata: Qwen3.8-27B, the Apache 2.0 vision-language model Q​wen published on 2026-08-05. So the architecture is Q​wen's, the weights are M​iniMax's fine-tune, and the packaging — NVFP4 rather than BF16 — is M​iniMax optimising for deployment rather than for redistribution fidelity. The card says the FP4 weights occupy about 19 GiB of GPU memory and that GPUs with native FP4 support run the NVFP4 weights directly while others fall back to bitsandbytes.

它確切來說是做什麼的

儲存庫裡有兩套工作流程，而它們完全不是同一回事。這是你下載 18 GB 之前最該先搞懂的一件事。

• 文字轉圖像 PE — 你能取得所有內容：微調後的權重、系統提示、任務提示、請求參數，以及可運作的用戶端。這條路徑現已完整且可執行。

• 影像編輯 PE — 你會拿到一個提示詞和一個訊息組裝範例。沒有權重。卡片直接寫明：「Image editing PE 仍在開發中；此版本僅提供其提示詞。」你需要將該提示詞交給自行選擇的視覺語言模型來執行。

文字轉圖像路徑會接收你的指令、擴充它，並回傳受結構描述約束、恰好包含兩個鍵的 JSON。針對測試提示「一隻橘色虎斑貓睡在陽光灑落的窗台上」的輸出，是一句關於溫暖照片與柔和午後光線的描述，再加上比例「3:2」。此結構描述很嚴格，會將 additionalProperties 設為 false，而 ratio 欄位是十個值的列舉：1:1、5:4、4:3、3:2、16:9、21:9、4:5、3:4、2:3、9:16。沒有自由文字的比例，輸出中沒有「auto」詞元，模型也無法用這十個字串以外的任何內容作答。這是刻意的設計選擇，且直接對應到父模型公布的尺寸表；該表正好列出這些比例，像素尺寸從 1:1 的 4096×4096 到 21:9 的 6272×2688。

比例決策不是憑感覺。已釋出的任務提示詞花了數百字說明一套嚴格的優先順序：明確指定的最終畫布最優先，其次是所要求的交付物（海報是直式、簡報頁首是橫式、頭像是正方形），再來是主體的主要構圖，最後才是可選的背景脈絡——並明確警告：「一般的房間、公園、森林、天空或電影感背景，不得讓聚焦的主體變成橫向。」提示詞也禁止自行臆造平台：你不能擅自假設是 Instagram。它還區分描繪場景與傳播成品——儀表板、海報、包裝、技術圖解或多格合成圖，會被建構為資訊架構，而不是一個包含所要求物件的場景。對任何看過影像模型把「季度業績簡報投影片」變成會議室照片的人來說，這項區分正是本次釋出內容有趣的地方。

為什麼要花 27B 參數在一段說明文字上？

答案就在父模型的說明文件中，而且比提示詞擴展器本身更有意思。MiniMax-H3-Image 是一個 33.1B 的稠密單流擴散 Transformer，50 層、隱藏大小 5376、56 個注意力頭，從一個完整 Qwen3-VL 文字編碼器的第 50 層讀取隱藏狀態；該編碼器原封不動地繼承自 MiniMax-H3。那個文字編碼器有 66.7 GB，且不在影像儲存庫中——你要另外從影片模型儲存庫下載。該檢查點經蒸餾後只需六個取樣步驟，且不使用無分類器引導，而 API 會強制執行這點：任何 num_inference_steps 只要不是 6，以及任何 guidance_scale 只要不是 1，都會回傳 HTTP 400，而 negative_prompt 會直接被拒絕。沒有任何旋鈕可以調整。

When there is no dial, the prompt is the control surface. That is the design logic connecting the two repositories: a six-step distilled sampler with no negative prompt has very little room to interpret a short instruction, so M​iniMax moved the interpretation into a separate model that can think before it writes. The hosted equivalent is an API endpoint called H3-Image-Context-IR, which does the same job server-side — accepts up to 32,000 characters of prompt and one to sixteen reference images, returns a revised prompt and a chosen ratio, and bills by converting the expansion model's input, output and thinking tokens into text tokens. MiniMax-H3-Image-PE is that endpoint's local escape hatch, and the parent card recommends exactly that usage: run the PE, save its expanded_caption into the generation call, and skip paying for enhancement twice.

The request parameters released with the PE are worth reading as a statement of intent rather than as tuning advice. Temperature 1.0, top_p 0.95, top_k 20, a 4096-token budget that includes thinking, and reasoning_effort set to "xhigh", with thinking enabled and preserved. M​iniMax is running its caption writer in maximum-deliberation mode and constraining only the final serialisation, not the reasoning. Serving that means SGLang 0.5.23 or newer with XGrammar 0.2.7 or newer, the q​wen3 reasoning parser, and constrained JSON decoding — or the Diffusers modular-blocks route the card documents. One caveat on that second route: the blocks repository the card points at, diffusers/minimax-h3-image-pe-blocks, returned HTTP 401 when checked on 2026-11-11, so it was not publicly readable at the time of writing. The card's Diffusers section may well be ahead of the artifact it depends on.

授權條款才是真正能阻止人們的部分

MiniMax-H3-Image-PE 並未單獨授權。模型卡指出，它「屬於 MiniMax-H3-Image 的一部分，並因此應受」《MiniMax-H3-Image Community License Agreement》規範，而授權檔案中有一行日期寫著「發行日期／授權日期：2026 年 11 月 10 日」。請仔細閱讀這一點，因為這表示提示詞擴充器的授權日期與父圖像模型的發布日期相隔八天，而授權本身卻是一份同時規範兩者的文件。

有三項條款比其他條款更重要，而且它們都不是常見的「開放權重」故事：

• 領土——授權僅在「適用領土」內適用，該領土定義為全球不包括歐洲聯盟、英國、大韓民國及美利堅合眾國。此為領土排除，而非使用限制，且此規定載於定義中，而非埋沒於附錄。

• 商業規模 — 若您或您關係企業的年化營收超過 100 萬美元，第 IV 條要求須另行取得事前書面授權，或若您有向第三方提供託管服務（無論規模）。第二項條件並無營收門檻。

• 介面標示——同一節要求您在使用本著作之任何商業產品或服務的使用者介面上，顯著顯示「MiniMax-H3-Image」。

此外還有一項再散布條件，要求提供含有特定著作權標示行的 NOTICE 檔案，以及第 V 節的義務：須維持防護措施，避免產生違規輸出，並設有通報機制及對善意通報採取行動的責任。對 2026 年的社群授權而言，這些都不算罕見。真正會讓許多團隊決定這 18 GB 的下載是否還值得開始的，是地域排除條款——而它也是最可能被漏看的條款，因為 Hugging Face 模型頁面只以「其他」加上一個連結來呈現它。

尚未可知的事物

把界線講清楚，比把它填滿更有用。

• 沒有基準測試數據。兩份模型卡都未公布任何一項分數。母模型僅提出質性主張——「開源模型中指令遵循能力最強」、「業界最先進」——而且截至 2026-11-11，未找到對 MiniMax-H3-Image 或其提示詞擴展器的獨立評估。沒有 Elo，沒有偏好勝率，也沒有與託管的 Context-IR 端點相比的擴展品質比較。

• 沒有證據顯示編輯路徑可行。編輯提示詞設想周到且異常具體——它要求呼叫方提供每張參考影像化簡後的 W:H 比例，要求模型絕不要從預覽圖估計來源比例，並指示模型建構預期的目標，而非假裝看過隱藏結果。但目前沒有為它發布任何權重，因此沒有人測量過一般 VLM 是否會像微調模型理應能做到的那樣遵循它。

• No announcement, therefore no stated intent. Whether this is a permanent commitment or a byproduct of the image launch is not stated anywhere. The repository has no roadmap, no model index entry, and no discussion from M​iniMax explaining why the expander was split out rather than folded into the image checkpoint.

• 下載數字太小，難以從中看出端倪。MiniMax-H3-Image 在 2026-11-11、發布後八天顯示 667 次下載和 136 個讚。PE 模型則顯示 0 次下載和 7 個讚。對於一個才成立一天的儲存庫來說，這只是下限，不是定論——但這確實意味著還沒有人發表實測報告。

如果你真的在打造某個東西，這適合放在哪裡

實務上的問題不是「這是不是一個好模型」——目前沒有人能回答——而是「採用它會讓我承諾什麼」。具體來說有三件事。你得承諾執行兩個模型而不是一個：一個 18 GB 的 expander，加上一個 68.65 GB 的影像檢查點，再加上一個 66.7 GB 的文字編碼器；如果你跑 MiniMax-H3，這個編碼器你很可能已經有了。你得承諾固定的 4096 個 token 解碼預算（包含推理）；而 API 文件把請求逾時設為 600 秒，這不是你能不加留意就藏在互動式 UI 後面的延遲。而且你得承諾採用一份地域範圍排除四個主要市場的授權條款。

The mitigation for the first two is the same one that applies to any model you have not measured yourself: put it behind a routing layer rather than into a hardcoded call path. This is the case routing is built for — the expander is a text-generation model behind an OpenAI-compatible /v1/chat/completions endpoint, so it can sit alongside everything else on one key, and a routing rule can send long or structured prompts to it while a shorter path handles the rest. If the expansion turns out to degrade a particular workload, failover moves that traffic without a redeploy. That is also the honest way to evaluate an unproven model: run it on a slice, keep the old path warm, and let the comparison decide. One clarification worth stating plainly, since it is easy to assume otherwise — MiniMax-H3-Image and MiniMax-H3-Image-PE are not on OrcaRouter's catalogue as of 2026-11-11, so this is a self-hosted evaluation, not a call we can make for you. What the catalogue does carry is M​iniMax's H3 video model and the M3.x text line at provider list price with zero markup passed through, which is the same arithmetic that applies the day an image endpoint arrives.

For the third — the territory clause — no amount of engineering helps. If you are in the EU, the UK, Korea or the US, the license does not grant you the rights it grants everyone else, and M​iniMax's own remedy is to contact them for a separate grant "based on robust controls and guardrails." That is a business conversation, and it should happen before the download, not after.

有什麼好看

Two signals would turn this from a repository into a release. The first is an announcement: a blog post, a model-index entry, a mention on M​iniMax's news feed. Its absence is currently the most informative thing about the launch, and its arrival would clarify whether the PE is meant to be the default local path or a stopgap. The second is the Diffusers blocks repository becoming publicly reachable — right now it returns 401, and until it resolves, the "run it locally through Diffusers" route documented on the card is a claim rather than a path.

In between, the thing to actually read is the task prompt. It is a few hundred lines of plain English describing how to decide what an image is for, and it is the most concrete artefact M​iniMax has published about how it thinks image prompts should work. You can learn more from it in ten minutes than from any benchmark table, and unlike the weights, it costs nothing to read.