# Full model list

> 124 callable models — 56 text (10 vendors + the platform's own xiaozhanAI), 23 image, 38 video, generated from the pricing registry

Source: https://qd.zsopc.com/en/docs/reference/models

The platform currently offers **124 callable models**: **56 text chat models** across **10 vendors** plus the platform's own `xiaozhanAI`, plus **23 image** (including the platform's own `xiaozhanimage`) and **38 video** generators, and 6 music / voice / avatar SKUs (last section on this page). Text pricing is in credits per 1M tokens; images bill per image and video per second.

> This page is **generated from the same pricing registry the gateway bills against**, so every id here is callable and every price is the one you are charged.
>
> ⚠️ The 124 here and the 125 on the [model catalogue](https://qd.zsopc.com/models) count different things — this is not drift: the plaza additionally shows not-yet-launched placeholder cards (badged "coming soon"), while this page lists only ids you can call **today**. Some SKUs that are not on the plaza (API-only, or temporarily not shown) appear on neither. Live data: `GET https://qd.zsopc.com/api/v1/models/public`, or the [model catalogue](https://qd.zsopc.com/models).

## Reading the tables

| Column | Meaning |
|---|---|
| capability `tools` | function calling |
| capability `vision` | image input |
| capability `web_search` | built-in web search tool |
| capability `image_gen` / `video_gen` | generate images / video inside a conversation |
| capability `thinking` | reasoning / thinking mode |
| capability `json` | enforced JSON output |
| `cached` | the price when input hits the prompt cache |

> **Video analysis in web chat**: in the AI chat on qd.zsopc.com, a text card that cannot read video can still take an upload (up to 300 MB). The model decides whether to open the cloud sandbox and extracts frames there with ffmpeg. It sees the picture and any text in it, and cannot hear sound. There is no separate charge: pictures bill as the model's tokens, and using the sandbox bills the usual container fee. This is only in web chat, not in the API; `supports_video_analysis` in `/api/v1/models/public` says which cards have it.

> **Cached pricing needs no opt-in**: when upstream usage reports `cached_tokens`, the gateway settles that portion of the input at the cached rate automatically — clients don't (and can't) enable it explicitly. Models without a `cached` figure bill all input at the Input price.

> **Only Anthropic Claude, `xiaozhanAI` and the `-bd` line charge for cache *writes*** (the `xiaozhanAI` and `-bd` write rates are in their own sections). The first time Claude writes a prefix into the cache, those tokens bill at **Input × 1.25** (`cache_creation_input_tokens`); later hits bill at the `cached` rate above (Input × 0.1). The cache lives about **5 minutes**, refreshed on every hit — if no follow-up turn reuses the prefix within that window, the write premium buys nothing. Every other vendor (the other OpenAI cards / xAI / Alibaba / Zhipu / Moonshot / DeepSeek / MiniMax / Tencent / Volcengine) only has a hit discount and does not bill writes separately.

> **Limited-time promotion**: the whole Anthropic Claude line is charged at **40% off** (list price × 0.6) and the whole OpenAI GPT text line (except the `-bd` line) at **70% off** (× 0.3), and `gemini-4-argon` at **50% off** (× 0.5), with cached rates discounted equally. The tables show list prices; actual billing always uses the discounted rate, for both API-key calls and in-platform conversations. Discounted models carry an "X% OFF" badge on the model catalogue.

> **Long-context tier**: for `grok-4.5` / `grok-4.6` / `grok-4.7`, a prompt of **≥200K tokens** bills input, cache-hit and output all at **2× the table price**. That is xAI's whole-turn doubling for very long requests, passed straight through (verified to the cent on 2026-08-23). The trigger is the total prompt size of the turn, independent of how much was generated.
>
> `gpt-6` / `gpt-6-astra` / `gpt-6-sol` / `gpt-6-luna` / `gpt-6.1-sol` / `gpt-6.1-terra` / `xiaozhanAI` and the seven `-bd` text cards work the same way: a prompt of **≥272K tokens** bills the whole turn at the long-context tier — input and cache-hit at **2×** the table price, output at **1.5×** — which is OpenAI's official whole-request repricing for very long prompts, passed straight through. No other model has this tier.

> **Fast tier**: the Fast-capable GPT cards (the ones with a "Fast" badge on the model catalogue) accept `"service_tier": "priority"` in the request body for the faster tier, billed at the **Fast price** — for `gpt-6.1-sol` / `gpt-6.1-terra` / `gpt-6-sol` / `gpt-6-luna` that matches OpenAI's official Fast pricing, **2×** the table prices (long-context tier and cached rates included); the other Fast-capable GPT cards are **1.5×**. Promotions still apply on top. The "Fast N×" badge on a model-catalogue card is that card's multiplier. If the fast lane is temporarily unavailable the turn is served — and billed — at the standard tier instead of failing. `xiaozhanAI` is not on this list: it always runs on the official Fast tier, and its table price already is the Fast price (see its section). The `-bd` line has no Fast tier.

> **Server-side tools bill per call**: when a grok model runs web search / tools, the upstream charges **2.022 points per tool call** on top of tokens. The count comes from the upstream's `usage.num_server_side_tools_used` and is written to the `server_tool_calls` column of every usage-log row, so you can re-derive your bill. Turns that use no tools incur no such charge. `xiaozhanAI` and the `-bd` line have their own per-search web-search price, see their sections.

## Drawing and the cloud sandbox

The text models below can draw pictures and can use a cloud sandbox to make Word / PowerPoint / Excel / PDF files, both in [AI Chat](https://qd.zsopc.com/chat) and over the API. The model decides from what you say: ask for a poster and it draws one, ask for a deck and it makes one, just chat and it answers in text.

| Vendor | Drawing | Cloud sandbox |
|---|---|---|
| Anthropic | 13 of the 14 (all but `claude-haiku-5-5`) | 13 of the 14 (all but `claude-haiku-5-5`) |
| OpenAI | the pool GPT cards draw by themselves (the ones marked `image_gen` in the tables); the 7 `-bd` cards draw through the platform | 16: `gpt-5.6-*`, `gpt-6*` and their `-bd` versions (not `codex-auto-review`) |
| xiaozhanAI | draws by itself | OpenAI-hosted container |
| Google | `gemini-3.8-flash`, `gemini-4-argon`, `gemini-3.8-flash-nx`, `gemini-4-argon-nx` | `gemini-3.8-flash`, `gemini-4-argon`, `gemini-3.8-flash-nx`, `gemini-4-argon-nx` |
| xAI | over the API `grok-4.6`; in AI Chat it draws when you turn on the 生图 (image) switch in the composer | over the API `grok-4.6` and `grok-4.7` (no sandbox in AI Chat) |
| Zhipu GLM | all 2 | all 2 |
| DeepSeek | `deepseek-v4-pro`, `deepseek-v4.1-flash` | `deepseek-v4-pro`, `deepseek-v4.1-flash` |
| Moonshot Kimi | all 5 | `kimi-k3`, `kimi-k3-1013` |
| MiniMax | `MiniMax-M2.7` | — |
| Alibaba Qwen | all 2 | all 2 |

> - **Billing**: tokens bill at the model's prices as usual. Cards that draw through the platform also pay, per image, the current price of the image card [GPT-Image 2.5 Flare](https://qd.zsopc.com/models/gpt-image-2.5-flare). The cloud sandbox costs **194.112 points** per conversation per **20 minutes**, charged only when a command actually ran in it; more files within the same 20 minutes cost nothing extra.
> - **AI Chat**: finished files stay in the conversation as file cards you can preview and download. Images and files uploaded in the conversation, and earlier pictures and files, are placed in the sandbox so the model can read and change them (up to 12, 20 MB each). Turns on the Fast tier, and turns when the balance cannot cover one container fee, do not use the sandbox; knowledge-base questions and model comparison have neither feature.
> - **API**: as with OpenAI, the tools are used only when the request declares them: `/v1/responses` takes `image_generation`, `shell` or `code_interpreter`; `/v1/chat/completions` takes `image_generation` only (with `stream: true`; the picture comes as `![](link)` in the answer). Without the tools it is the plain model. See [Cloud sandbox](../guides/cloud-sandbox.md) and [Image editing](../guides/image-editing.md). A turn whose balance cannot cover one container fee does not use the sandbox; the model answers in text.


## Anthropic (14)

| Model ID | Name | Context | Input (积分/1M) | Output (积分/1M) | Capabilities |
|---|---|---|---|---|---|
| `claude-fable-5` | Claude Fable 5 | 200K | 4044 (cached 404.4) | 20220 | tools / vision / web_search / thinking / json |
| `claude-fable-5-1` | Claude Fable 5.1 | 1M | 4044 (cached 101.1) | 20220 | tools / vision / web_search / thinking / json |
| `claude-haiku-4-5-20251001` | Claude Haiku 4.5 | 200K | 404.4 (cached 40.44) | 2022 | tools / vision / web_search / thinking / json |
| `claude-haiku-5-5` | Claude Haiku 5.5 | 1M | 40.44 (cached 4.044) | 202.2 | tools / vision / thinking / json |
| `claude-opus-4-5-20251101` | Claude Opus 4.5 | 200K | 2022 (cached 202.2) | 10110 | tools / vision / web_search / thinking / json |
| `claude-opus-4-6` | Claude Opus 4.6 | 1M | 2022 (cached 202.2) | 10110 | tools / vision / web_search / thinking / json |
| `claude-opus-4-7` | Claude Opus 4.7 | 1M | 2022 (cached 202.2) | 10110 | tools / vision / web_search / thinking / json |
| `claude-opus-4-8` | Claude Opus 4.8 | 1M | 2022 (cached 202.2) | 10110 | tools / vision / web_search / thinking / json |
| `claude-opus-5` | Claude Opus 5 | 1M | 2022 (cached 202.2) | 10110 | tools / vision / web_search / thinking / json |
| `claude-opus-5-5` | Claude Opus 5.5 | 1M | 1617.6 (cached 80.88) | 8088 | tools / vision / web_search / thinking / json |
| `claude-sonnet-4-5-20250929` | Claude Sonnet 4.5 | 200K | 1213.2 (cached 121.32) | 6066 | tools / vision / web_search / thinking / json |
| `claude-sonnet-4-6` | Claude Sonnet 4.6 | 1M | 1213.2 (cached 121.32) | 6066 | tools / vision / web_search / thinking / json |
| `claude-sonnet-5` | Claude Sonnet 5 | 1M | 808.8 (cached 80.88) | 4044 | tools / vision / web_search / thinking / json |
| `claude-sonnet-5-5` | Claude Sonnet 5.5 | 1M | 111.323232 (cached 5.58072) | 556.61616 | tools / vision / web_search / thinking / json |

## OpenAI (17)

| Model ID | Name | Context | Input (积分/1M) | Output (积分/1M) | Capabilities |
|---|---|---|---|---|---|
| `codex-auto-review` | Codex Auto Review | 272K | 80.88 (cached 8.088) | 485.28 | tools / vision / web_search / image_gen / thinking / json |
| `gpt-5.6-luna` | GPT-5.6 Luna | 1M | 80.88 (cached 8.088) | 485.28 | tools / vision / web_search / image_gen / thinking / json |
| `gpt-5.6-sol` | GPT-5.6 Sol | 1M | 2022 (cached 202.2) | 12132 | tools / vision / web_search / image_gen / thinking / json |
| `gpt-5.6-terra` | GPT-5.6 Terra | 1M | 2022 (cached 202.2) | 12132 | tools / vision / web_search / image_gen / thinking / json |
| `gpt-6` | GPT-6 | 1.05M | 4044 (cached 404.4) | 20220 | tools / vision / web_search / image_gen / thinking / json |
| `gpt-6-astra` | GPT-6 Astra | 1.05M | 4044 (cached 404.4) | 20220 | tools / vision / web_search / image_gen / thinking / json |
| `gpt-6-luna` | GPT-6 Luna | 1.05M | 40.44 (cached 4.044) | 202.2 | tools / vision / web_search / image_gen / thinking / json |
| `gpt-6-sol` | GPT-6 Sol | 1.05M | 808.8 (cached 80.88) | 4044 | tools / vision / web_search / image_gen / thinking / json |
| `gpt-6.1-sol` | GPT-6.1 Sol | 1.05M | 808.8 (cached 40.44) | 4044 | tools / vision / web_search / image_gen / thinking / json |
| `gpt-6.1-terra` | GPT-6.1 Terra | 1.05M | 404.4 (cached 20.22) | 2022 | tools / vision / web_search / image_gen / thinking / json |
| `gpt-5.6-luna-bd` | GPT-5.6 Luna-bd | 1.05M | 80.88 (cached 8.088) | 485.28 | tools / vision / web_search / thinking / json |
| `gpt-5.6-sol-bd` | GPT-5.6 Sol-bd | 1.05M | 1617.6 (cached 161.76) | 8088 | tools / vision / web_search / thinking / json |
| `gpt-5.6-terra-bd` | GPT-5.6 Terra-bd | 1.05M | 808.8 (cached 80.88) | 4852.8 | tools / vision / web_search / thinking / json |
| `gpt-6-astra-bd` | GPT-6 Astra-bd | 1.05M | 4044 (cached 404.4) | 20220 | tools / vision / web_search / thinking / json |
| `gpt-6-luna-bd` | GPT-6 Luna-bd | 1.05M | 40.44 (cached 4.044) | 202.2 | tools / vision / web_search / thinking / json |
| `gpt-6-sol-bd` | GPT-6 Sol-bd | 1.05M | 808.8 (cached 80.88) | 4044 | tools / vision / web_search / thinking / json |
| `gpt-6.1-sol-bd` | GPT-6.1 Sol-bd | 1.05M | 808.8 (cached 40.44) | 4044 | tools / vision / web_search / thinking / json |

> - The `-bd` line (`gpt-5.6-luna-bd` / `gpt-5.6-sol-bd` / `gpt-5.6-terra-bd` / `gpt-6-astra-bd` / `gpt-6-luna-bd` / `gpt-6-sol-bd` / `gpt-6.1-sol-bd`): a separate route (the bd line) billed at OpenAI's official Standard list price — **no limited-time promotion and no Fast tier**, so the table price is what you are charged; a prompt of ≥272K tokens bills the whole turn at the long-context tier (see "Long-context tier" at the top of this page).
> - On the `-bd` line, **cache writes** bill at **Input × 1.25** (`cache_write_tokens` in usage) and cache hits at the cached rate; **web search** (`web_search` tool) bills **4.044 points per search**, counted from the upstream's reported `tool_usage.web_search.num_requests` and written to the `server_tool_calls` column of the usage log.
> - The `-bd` line serves OpenAI Chat and OpenAI Responses. Reasoning effort on Responses (`reasoning.effort`) accepts `low` / `medium` / `high` / `xhigh` / `max`; on Chat (`reasoning_effort`) the highest is `xhigh` (Chat requests that carry function tools follow the Responses rules and accept `max`). `gpt-5.6-*-bd`, `gpt-6-sol-bd` and `gpt-6-luna-bd` also accept `none`. Function tools, image input, file input (PDF) and web search (with citations) are supported; **in-conversation image generation (the `image_generation` tool) is not**. `/v1/responses/compact` is not supported.

## xiaozhanAI (1)

The platform's own-brand chat model, served straight from the official OpenAI API (no account pool or relay). **Every call runs on OpenAI's official Fast tier** (about 1.7× the standard output speed); the table shows the Fast price, and a `service_tier` in the request does not change the tier or the bill. Not part of the limited-time promotion.

| Model ID | Name | Context | Input (积分/1M) | Output (积分/1M) | Capabilities |
|---|---|---|---|---|---|
| `xiaozhanAI` | xiaozhanAI | 1.05M | 121.32 (cached 12.132) | 606.6 | tools / vision / web_search / image_gen / thinking / json |

> - **Web search** (`web_search` tool) bills **8.088 points per search**, counted from OpenAI's reported `tool_usage.web_search.num_requests` (a search counts once; opening a page is not charged separately) and written to the `server_tool_calls` column of the usage log.
> - **Image generation** (`image_generation` tool) bills **170.6568 points per image**, independent of tokens.
> - **Cache writes** bill at **151.65 points / 1M** (`cache_write_tokens` in usage); cache hits bill at the cached rate.
> - **Cloud sandbox** (produces Word / Excel / PPT / PDF files): `/v1/responses` with the `shell` or `code_interpreter` tool; download files via `/v1/containers/{id}/files/{file_id}/content`. Container fee: **194.112 points** per container per 20 minutes. Usage and limits: [Cloud sandbox](../guides/cloud-sandbox.md).
> - Protocols: OpenAI Chat and OpenAI Responses (the cloud sandbox is Responses-only). `background: true` is not supported (use `stream: true` for long requests).

## Google (6)

| Model ID | Name | Context | Input (积分/1M) | Output (积分/1M) | Capabilities |
|---|---|---|---|---|---|
| `gemini-3.1-pro-low` | Gemini 3.1 Pro | 2M | 808.8 (cached 80.88) | 4852.8 | tools / vision / web_search / thinking / json |
| `gemini-3.8-flash` | Gemini 3.8 Flash | 1M | 303.3 (cached 30.33) | 1516.5 | tools / vision / thinking / json |
| `gemini-4-argon` | Gemini 4 Argon | 1M | 808.8 (cached 40.44) | 4044 | tools / vision / thinking / json |
| `gemini-3.1-pro-preview-1020` | Gemini 3.1 Pro Preview 1020 | 1M | 808.8 (cached 80.88); >200K 1617.6 (cached 161.76) | 4852.8; >200K 7279.2 | tools / vision / web_search / thinking / json |
| `gemini-3.8-flash-nx` | Gemini 3.8 Flash-nx | 1M | 303.3 (cached 30.33) | 1516.5 | tools / vision / thinking / json |
| `gemini-4-argon-nx` | Gemini 4 Argon-nx | 1M | 808.8 (cached 40.44) | 4044 | tools / vision / thinking / json |

> - `gemini-4-argon`: the Gemini 4 flagship reasoning model, with thinking levels `low` / `medium` (default) / `high` (Gemini native `thinkingConfig.thinkingLevel`). Supports function calling, image input and `responseSchema` structured output; no web search (`googleSearch`).
> - `gemini-3.8-flash-nx` / `gemini-4-argon-nx`: a separate relay line (nx) serving models under the same names as `gemini-3.8-flash` / `gemini-4-argon`, billed at Google's list price per token (no discount). Supports function calling, image input and `responseSchema` structured output; no web search (`googleSearch`); `gemini-4-argon-nx` takes the same thinking levels as `gemini-4-argon`. Cache hits are not reported, so input is billed in full. This line has limited capacity: under high concurrency responses get noticeably slower or fail.
> - `gemini-3.1-pro-preview-1020`: a separate relay line (1020), billed at Google's list price per token. Supports Google Search grounding (`googleSearch`), `responseSchema` structured output, function calling and image input; it does not report cache hits, so input bills at the full rate.

## xAI (3)

| Model ID | Name | Context | Input (积分/1M) | Output (积分/1M) | Capabilities |
|---|---|---|---|---|---|
| `grok-4.5` | Grok 4.5 | 500K | 808.8 (cached 121.32) | 2426.4 | tools / vision / thinking / json |
| `grok-4.6` | Grok 4.6 | 500K | 808.8 (cached 202.2) | 2426.4 | tools / vision / thinking / json |
| `grok-4.7` | Grok 4.7 | 500K | 808.8 (cached 202.2) | 2426.4 | tools / vision / thinking / json |

## Zhipu GLM (4)

| Model ID | Name | Context | Input (积分/1M) | Output (积分/1M) | Capabilities |
|---|---|---|---|---|---|
| `glm-5.3` | GLM-5.3 | 1M | 480 (cached 120) | 1680 | tools / thinking / json / open-source |
| `glm-5.3-1022` | GLM-5.3-1022 | 1M | 480 (cached 120) | 1680 | tools / thinking / json / open-source |
| `glm-5.3-flash` | GLM-5.3-Flash | 1M | 48 (cached 13.8) | 168 | tools / vision / thinking / json / open-source |
| `glm-5.3-flash-1022` | GLM-5.3-Flash-1022 | 1M | 48 (cached 13.8) | 168 | tools / vision / thinking / json / open-source |

> - The `-1022` line (`glm-5.3-1022` / `glm-5.3-flash-1022`): a separate route billed at Zhipu's official price (`glm-5.3-flash-1022` uses the standard price, not the limited-time discount). Both serve OpenAI Chat, Responses and Anthropic Messages; thinking is always on and cannot be turned off (`thinking.type=disabled` returns 400), and `reasoning_effort` accepts `low` / `high` / `max`. `glm-5.3-1022` is text-only; `glm-5.3-flash-1022` accepts images. `/v1/responses/compact` is not supported.

## DeepSeek (3)

| Model ID | Name | Context | Input (积分/1M) | Output (积分/1M) | Capabilities |
|---|---|---|---|---|---|
| `deepseek-v4-flash` | DeepSeek V4 Flash | 1M | 120 (cached 2.4) | 480 | tools / thinking / json |
| `deepseek-v4-pro` | DeepSeek V4 Pro | 128K | 540 (cached 18) | 1620 | tools / thinking / json |
| `deepseek-v4.1-flash` | DeepSeek V4.1 Flash | 1M | 120 (cached 2.4) | 480 | tools / vision / thinking / json / open-source |

> - **Every DeepSeek card follows DeepSeek's official peak / off-peak prices**: the table shows the peak price. Peak is Beijing time Mon–Fri (excluding Chinese public holidays) 9:00–12:00 and 14:00–18:00; everything else (including weekends and public holidays all day) is off-peak, when input, cache hits and output are all half price (e.g. `deepseek-v4-flash` and `deepseek-v4.1-flash` off-peak 60 (cached 1.2) / 240, `deepseek-v4-pro` off-peak 270 (cached 9) / 810). Decided by the time the request reaches the gateway.

## Moonshot Kimi (5)

| Model ID | Name | Context | Input (积分/1M) | Output (积分/1M) | Capabilities |
|---|---|---|---|---|---|
| `kimi-k2.5` | Kimi K2.5 | 256K | 242.64 (cached 40.44) | 1011 | tools / thinking / json / open-source |
| `kimi-k2.6` | Kimi K2.6 | 256K | 242.64 (cached 40.44) | 1011 | tools / thinking / json / open-source |
| `kimi-k2.7-code` | Kimi K2.7 Code | 256K | 242.64 (cached 40.44) | 1011 | tools / thinking / json / open-source |
| `kimi-k3` | Kimi K3 | 256K | 1200 (cached 120) | 6000 | tools / thinking / json / open-source |
| `kimi-k3-1013` | Kimi K3 1013 | 256K | 1200 (cached 120) | 6000 | tools / thinking / json / open-source |

> ⚠️ **`kimi-k3` / `kimi-k3-1013` parameter limits** (fixed by the upstream; any other value returns 400, same on `/v1/chat/completions` and `/v1/messages`): `temperature` must be 1, `top_p` must be 0.95, `presence_penalty` / `frequency_penalty` must be 0, `n` must be 1, `stop` takes at most 5 entries, `top_logprobs` is not supported; with thinking enabled, `tool_choice` cannot name a specific tool. Simply leave these parameters out. The model thinks before answering, so don't set `max_tokens` too low (the reply comes back empty); on the Anthropic protocol (`/v1/messages`) only **streamed** responses report cache hits and get the cached-input price.

## MiniMax (1)

| Model ID | Name | Context | Input (积分/1M) | Output (积分/1M) | Capabilities |
|---|---|---|---|---|---|
| `MiniMax-M2.7` | MiniMax M2.7 | 200K | 121.32 (cached 24.26) | 606.6 | tools / json |

## Alibaba Qwen (2)

| Model ID | Name | Context | Input (积分/1M) | Output (积分/1M) | Capabilities |
|---|---|---|---|---|---|
| `qwen3.8-max-cjx` | Qwen3.8-Max-cjx | 1M | 720 (cached 90) | 2160 | tools / vision / thinking / json |
| `qwen3.8-omni-flash` | Qwen3.8-Omni-Flash | 1M | 48 (cached 6) | 162 | tools / vision / thinking / json |

> - `qwen3.8-max-cjx`: a separate route (cjx) billed at the vendor's official list price; it serves OpenAI Chat and Responses. On Responses, `previous_response_id` and `/v1/responses/compact` are not supported (send the full history each turn), and the `web_search` tool does not actually search. Thinking tokens are not capped by `max_tokens` and are billed as the upstream reports them.

## Baidu ERNIE (1)

| Model ID | Name | Context | Input (积分/1M) | Output (积分/1M) | Capabilities |
|---|---|---|---|---|---|
| `ernie-5.5-flash-preview` | ERNIE 5.5 Flash Preview | 1M | 120 (cached 12) | 480 | tools / thinking / json |

> - **Billed at Baidu's official peak / off-peak prices**: the table shows the peak (official list) price. Peak is 8:00–22:00 Beijing time every day and off-peak is 22:00 to 8:00 the next morning (weekends and public holidays use the same hours); off-peak, input, cache hits and output are all half price: 60 (cached 6) / 240. Decided by the time the request reaches the gateway.
> - Reasoning cannot be turned off: every turn thinks first. `thinking`, `enable_thinking` and `reasoning_effort` are accepted but do not switch it off, and thinking tokens bill as output; don't set `max_tokens` too small (the answer comes back empty). Up to 1M output tokens. Text only: a request with an image returns 400.

## Image generation models (23)

Billed **per image** through `POST /v1/images/generations`. See [Image / video APIs](./media-apis.md).

### xiaozhanimage

The platform's own image model, served directly by the official OpenAI API (no pool or relay).

| Model ID | Name | Price |
|---|---|---|
| `xiaozhanimage` | xiaozhanimage | 30/image |

> - A flat **30 credits per image**, whatever the size or number of reference images; `quality` is fixed to `medium` by the platform (a value in the request is ignored).
> - **`size` is honoured** (pixel width×height, e.g. `1024x1536`): both sides must be multiples of 16 and the long side ≤ 3840, up to `3840x2160`; `auto` also works.
> - Reference-image edits go through `POST /v1/images/edits` (or send `image` on a generations request), **at most 9 reference images per request**; `n` returns several images, billed per image actually returned.
> - `stream` / `partial_images` are not supported; `response_format` defaults to a URL, `b64_json` returns base64.

### ByteDance Seedream

| Model ID | Name | Price |
|---|---|---|
| `doubao-seedream-4-0-250828` | Seedream 4.0 | 12/image |
| `doubao-seedream-4-5-251128` | Seedream 4.5 | 15/image |
| `doubao-seedream-5-0-260128` | Seedream 5.0 | 13.2/image |
| `doubao-seedream-5-0-pro-260628` | Seedream 5.0 Pro | 18/image ≤ 2.61 MP, 36 above; first input reference image free, then 1.2 each |

### Google

| Model ID | Name | Price |
|---|---|---|
| `nano-banana` | Nano Banana | 18/image |
| `nano-banana-2` | Nano Banana 2 | 18/image |
| `nano-banana-2.1` | Nano Banana 2.1 | 18/image |
| `nano-banana-pro` | Nano Banana Pro | 18/image |

### OpenAI

| Model ID | Name | Price |
|---|---|---|
| `gpt-image-2` | GPT-Image 2 | 24/image |
| `gpt-image-2.5-flare` | GPT-Image 2.5 Flare | 12/image |
| `gpt-image-2.5-sunburst` | GPT-Image 2.5 Sunburst | 12/image |
| `gpt-image-2-bd` | GPT-Image 2-bd | tokens: text input 2022 / image input 3235.2 / image output 12132 per 1M |
| `gpt-image-2.5-flare-bd` | GPT-Image 2.5 Flare-bd | tokens: text input 2022 / image input 3235.2 / image output 12132 per 1M |
| `gpt-image-2.5-sunburst-bd` | GPT-Image 2.5 Sunburst-bd | tokens: text input 2022 / image input 3235.2 / image output 12132 per 1M |
| `gpt-image-2.6-flare` | GPT-Image 2.6 Flare | tokens: text input 2022 / image input 3235.2 / image output 12132 per 1M |
| `gpt-image-2.6-sunburst` | GPT-Image 2.6 Sunburst | tokens: text input 2022 / image input 3235.2 / image output 12132 per 1M |

> - The three `-bd` image cards (the bd line) bill **per token**, not per image, at OpenAI's official list rates: text input 2022 / image input 3235.2 / image output 12132 points per 1M tokens (reference images count as image input). **`size` is honoured** (both sides must be multiples of 16), and so is `quality` (`low` / `medium` / `high` / `auto`); reference-image edits go through `POST /v1/images/edits`.
> - `gpt-image-2.6-flare` / `gpt-image-2.6-sunburst` (GPT Image 2.6, served straight from OpenAI's official API) also bill **per token**, at OpenAI's official list rates with no discount: text input 2022 / image input 3235.2 / image output 12132 points per 1M tokens. **`size` is honoured** (both sides multiples of 16, long side ≤ 3840), and so is `quality` (`low` / `medium` / `high` / `xhigh` / `max` / `auto`); reference-image edits go through `POST /v1/images/edits` (up to 9 references). **They can make animated GIFs**: `output_format: "gif"`, with optional `frame_count` (2–16, default 9) and `frame_duration_ms` (20–2000 ms, default 200); `size` is then the size of each frame. A GIF can take one reference image (to animate it), but no `mask`. See [Image / video APIs](./media-apis.en.md), section "Animated GIFs".

### xAI Grok

| Model ID | Name | Price |
|---|---|---|
| `grok-imagine-image` | Grok Imagine (Image) | 28.31/image |
| `grok-imagine-image-quality` | Grok Imagine (Image · Quality) | 28.31/image |

### Kuaishou Kling

| Model ID | Name | Price |
|---|---|---|
| `tvod-kling-image-3.0` | Kling Image 3.0 | 20.22/image |
| `tvod-kling-image-3.0-omni` | Kling Image 3.0 Omni | 24.26/image |

### Alibaba Qwen

| Model ID | Name | Price |
|---|---|---|
| `qwen-image-3.0` | Qwen Image 3.0 | 10.8/image |
| `qwen-image-3.0-pro` | Qwen Image 3.0 Pro | 1K 15 / 2K 30 per image |

> `gpt-image-2`, `gpt-image-2.5-flare` and `gpt-image-2.5-sunburst` all accept `POST /v1/images/edits` for reference-image edits (sketch-to-image works the same way). **The upstream ignores `size`** — put the aspect ratio in the prompt; latency is 30–215s.
>
> Since 2026-07-08 the three nano entries are the only way in: older ids (`gemini-2.5-flash-image`, `gemini-3-pro-image-preview`, `gemini-3.1-flash-image(-preview)`, `tvod-nano-*`) keep working as aliases onto the canonical entries above.

## Video generation models (38)

Billed **per second** as an async task through `POST /v1/videos/generations`. Where tiers are listed (resolution, or operation for the subtitle SKU) you are billed at the tier you request. See [Image / video APIs](./media-apis.md).

### ByteDance Seedance

| Model ID | Name | Price |
|---|---|---|
| `doubao-seedance-2.0` | Seedance 2.0 | 480p 27.72144 / 720p 59.616 / 1080p 148.716 / 4k 303.264 per second (4K is rendered natively); token rate by delivery tier: 480p 2760/1M (no video input) 1680/1M (with video input), 720p 2760/1M (no video input) 1680/1M (with video input), 1080p 3060/1M (no video input) 1860/1M (with video input), 4k 1560/1M (no video input) 960/1M (with video input) |
| `doubao-seedance-2.0-fast` | Seedance 2.0 Fast | 480p 22.29768 / 720p 47.952 per second; token rate by delivery tier: 480p 2220/1M (no video input) 1320/1M (with video input), 720p 2220/1M (no video input) 1320/1M (with video input) |
| `doubao-seedance-2.0-fast-lite` | Seedance 2.0 Fast Lite | 720p 47.952 per second (720p only); **same price and billing formula as `doubao-seedance-2.0-fast`**, token rate by delivered tier: 720p 2220/1M (default) 1320/1M (with video input) |
| `doubao-seedance-2.0-lite` | Seedance 2.0 Lite | 720p 59.616 / 1080p 148.716 per second; billed by token = **token rate x tokens + upscale rate x delivered seconds**. Token rate 720p 5686.58/1M (no video input) 3461.4/1M (with video input), 1080p 6653.52/1M (no video input) 4049.97/1M (with video input). Upscale add-on: 720p 2.5/s, 1080p 5/s (charged on **delivered** seconds only, never on the reference video) |
| `doubao-seedance-2.0-mini` | Seedance 2.0 Mini | 480p 13.86072 / 720p 29.808 per second; token rate by delivery tier: 480p 1380/1M (no video input) 840/1M (with video input), 720p 1380/1M (no video input) 840/1M (with video input) |
| `doubao-seedance-2.0-mini-lite` | Seedance 2.0 Mini Lite | 720p 29.808 per second (720p only); **same price and billing formula as `doubao-seedance-2.0-mini`**, token rate by delivered tier: 720p 1380/1M (default) 840/1M (with video input) |
| `doubao-seedance-2.5` | Seedance 2.5 | 480p 40.3515 / 720p 90.72 / 1080p 224.532 per second; token rate by delivery tier: 4200/1M (no video input), 2520/1M (with video input); the **1080p tier** bills 4620/1M and 2760/1M |
| `doubao-seedance-2.5-lite` | Seedance 2.5 Lite | 720p 42.8515 / 1080p 95.72 per second; billed by token = **token rate x tokens + upscale rate x delivered seconds**. Token rate 4200/1M (no video input), 2520/1M (with video input). Upscale add-on: 720p 2.5/s, 1080p 5/s (charged on **delivered** seconds only, never on the reference video) |
| `seedance2.0-yd` | Seedance 2.0 YD | 480p 27.72144 / 720p 59.616 / 1080p 148.716 per second; text-to-video and multimodal reference (up to 9 reference images with `role: reference_image`, optionally up to 3 reference audio clips with `role: reference_audio`); **no first frame / first-last frame and no reference video**; a text prompt, resolution, a concrete ratio (16:9 / 4:3 / 1:1 / 3:4 / 9:16 / 21:9, no adaptive) and an integer duration are required; token rate by delivery tier: 480p 2760/1M, 720p 2760/1M, 1080p 3060/1M |

### Alibaba Wan

| Model ID | Name | Price |
|---|---|---|
| `wan3.0-video` | Wan 3.0 video | 480p 18 / 720p 36 / 1080p 72 per second |
| `wan3.0-video-prime` | Wan 3.0 video Prime | 480p 27 / 720p 54 / 1080p 108 per second |

> Both Wan 3.0 cards share one capability set: 2-30s, 480P/720P/1080P, 30fps, aspect `16:9 / 4:3 / 1:1 / 3:4 / 9:16 / adaptive`; text-to-video, image-to-video (first frame, first+last frame), and image / video / audio references. **Prime is the accelerated tier** (same quality, markedly faster) at 1.5x the standard rate. **The default delivery is 1080P** — pass `resolution` explicitly for a cheaper tier.

> ⚠️ **On rounds that carry a reference video, the input video's seconds are billed too.** This is the upstream's own basis: billed duration = input duration + output duration, and the rate above applies to both legs (upstream's price table heads that column "input and output unit price"). Reference videos are `[1,15]`s each and 15s total; we hold against that ceiling at submit time and refund the difference once upstream reports the real input duration. Reconcile against the top-level `input_video_duration_sec` in the response (`duration_sec` still means the delivered clip only, excluding the input). **Text-to-video and image-to-video rounds are unaffected** (image input is free upstream, and we don't charge for it either). See [Media APIs - how `usage` relates to billing](./media-apis.en.md#how-usage-relates-to-billing).

### Google

| Model ID | Name | Price |
|---|---|---|
| `tvod-veo-3.1` | Veo 3.1 | 720p 24.26 / 1080p 40.44 / 4k 121.32 per second |
| `tvod-veo-3.1-fast` | Veo 3.1 Fast | 720p 40.44 / 1080p 48.53 per second |
| `tvod-veo-3.1-lite` | Veo 3.1 Lite | 720p 20.22 / 1080p 32.35 per second |

### xAI Grok

| Model ID | Name | Price |
|---|---|---|
| `grok-imagine-video` | Grok Imagine Video | 283.08/second |
| `grok-imagine-video-1.5-preview` | Grok Imagine Video 1.5 | 586.38/second |

### Kuaishou Kling

| Model ID | Name | Price |
|---|---|---|
| `tvod-kling-video-4.0` | Kling 4.0 | 720p 54 / 1080p 72 / 2k 90 / 4k 180 per second |
| `tvod-kling-video-4.0-omni` | Kling 4.0 Omni | 720p 54 / 1080p 72 / 2k 90 / 4k 180 per second |
| `tvod-kling-video-3.0` | Kling 3.0 | 720p 20.22 / 1080p 32.35 / 4k 97.06 per second |
| `tvod-kling-video-3.0-omni` | Kling 3.0 Omni | 720p 20.22 / 1080p 32.35 / 4k 97.06 per second |
| `tvod-kling-video-o1` | Kling O1 | 720p 12.13 / 1080p 20.22 per second |

> Kling 4.0 / 4.0 Omni: integer durations of 3-30s, aspect `16:9 / 9:16 / 1:1 / 21:9`, native audio in every clip. The official price has not been published yet; these cards bill at Kling 3.0's official with-audio rates until it is, then follow the official price.

### MiniMax Hailuo

| Model ID | Name | Price |
|---|---|---|
| `minimax-h3` | Hailuo H3 | 768p 30 / 2k 48 per second |
| `minimax-h3-max` | Hailuo H3 Max | 480p 19.8 / 768p 30 per second |
| `minimax-h3-tx` | Hailuo H3 tx | 768p 30 / 720p 30 per second |
| `minimax-h3-fast-tx` | 海螺 H3 提速版 tx | 768p 24 / 720p 24 / 480p 15.6 per second |
| `minimax-h3-medium-tx` | 海螺 H3 满血版 tx | 768p 30 / 720p 30 per second |
| `minimax-h3-fast-ir-tx` | 海螺 H3 提速版带IR tx | 768p 24 / 720p 24 / 480p 15.6 per second |
| `minimax-h3-medium-ir-tx` | 海螺 H3 满血版带IR tx | 768p 30 / 720p 30 per second |

> The `-tx` cards run on the same line with the same parameters and capabilities; only the model version differs. `minimax-h3-tx` is the default version (the former id `minimax-h3-1014` still works). Fast (提速版) has faster inference, bills at the official MiniMax-H3-Highspeed price and also offers a 480p tier; Full (满血版) is full precision and bills at the official MiniMax-H3 price. "IR" (带IR) adds an upstream prompt-rewriting step at no extra charge. All five cards bill the same way (reference-video seconds are billed, up to 5 reference images free).

### Shengshu Vidu

| Model ID | Name | Price |
|---|---|---|
| `tvod-vidu-video-q3` | Vidu Q3 | 720p 18.2 / 1080p 28.31 / 4k 80.88 per second |
| `tvod-vidu-video-q3-ad` | Vidu Q3 Ads | 720p 18.2 / 1080p 28.31 / 4k 80.88 per second |
| `tvod-vidu-video-q3-drama` | Vidu Q3 Drama | 720p 18.2 / 1080p 28.31 / 4k 80.88 per second |
| `tvod-vidu-video-q3-pro` | Vidu Q3 Pro | 720p 18.2 / 1080p 28.31 / 4k 80.88 per second |
| `tvod-vidu-video-q3-turbo` | Vidu Q3 Turbo | 720p 18.2 / 1080p 28.31 per second |

### Happyhorse

| Model ID | Name | Price |
|---|---|---|
| `happyhorse-1.1-i2v` | Happyhorse 1.1 image-to-video | 720p 54 / 1080p 72 per second |
| `happyhorse-1.1-r2v` | Happyhorse 1.1 omni-reference | 720p 54 / 1080p 72 per second |
| `happyhorse-1.1-t2v` | Happyhorse 1.1 text-to-video | 720p 54 / 1080p 72 per second |

### Digital human / editing

| Model ID | Name | Price |
|---|---|---|
| `clip-compose` | Trend-style video edit | 48/call |
| `dh-avatar` | Digital-human video | 3.96/second |

## Music / voice / avatar creation (6)

These have their own endpoints and billing units. Full request/response shapes: [Image / video / music APIs](./media-apis.md) and [Digital human / editing APIs](./digital-human-apis.md).

| Model ID | Name | Price | Endpoint |
|---|---|---|---|
| `dh-avatar-create` | Create digital-human avatar | 220.11/avatar | `POST /v1/videos/generations` (avatar-creation task) |
| `suno-v6` | Suno V6 (latest) | 20.4/generation (2 songs) | `POST /v1/music/generations` |
| `suno-v6-mini` | Suno V6 Mini | 20.4/generation (2 songs) | `POST /v1/music/generations` |
| `suno-v6-wild` | Suno V6 Wild | 20.4/generation (2 songs) | `POST /v1/music/generations` |
| `voice-clone-pro` | Voice clone · create a voice | 40.44/voice | `POST /v1/audio/voice-clone` |
| `voice-tts-pro` | Text to speech | 0.1321/character | `POST /v1/audio/speech` |

## Cross-vendor protocol support

Different models support different client SDKs, and **the [compatibility matrix](./compatibility-matrix.md) is authoritative on what is actually reachable**. The catalog's `supported_protocols` field reflects only the **native / primary path**; many models are additionally reachable on other protocols through gateway translation (the response then carries an `X-Protocol-Translation` header). Native paths, counted from the registry that generated this page:

- **OpenAI Chat** (`/v1/chat/completions`) — all 57 text models
- **Anthropic Messages** (`/v1/messages`) — 18 models: `claude-fable-5`, `claude-fable-5-1`, `claude-haiku-4-5-20251001`, `claude-haiku-5-5`, `claude-opus-4-5-20251101`, `claude-opus-4-6`, `claude-opus-4-7`, `claude-opus-4-8`, `claude-opus-5`, `claude-opus-5-5`, `claude-sonnet-4-5-20250929`, `claude-sonnet-4-6`, `claude-sonnet-5`, `claude-sonnet-5-5`, `glm-5.3-1022`, `glm-5.3-flash-1022`, `kimi-k3`, `kimi-k3-1013`
- **OpenAI Responses** (`/v1/responses`) — 30 models: `claude-haiku-4-5-20251001`, `claude-opus-4-6`, `claude-opus-4-7`, `claude-opus-4-8`, `claude-sonnet-4-6`, `claude-sonnet-5`, `codex-auto-review`, `glm-5.3-1022`, `glm-5.3-flash-1022`, `gpt-5.6-luna`, `gpt-5.6-luna-bd`, `gpt-5.6-sol`, `gpt-5.6-sol-bd`, `gpt-5.6-terra`, `gpt-5.6-terra-bd`, `gpt-6`, `gpt-6-astra`, `gpt-6-astra-bd`, `gpt-6-luna`, `gpt-6-luna-bd`, `gpt-6-sol`, `gpt-6-sol-bd`, `gpt-6.1-sol`, `gpt-6.1-sol-bd`, `gpt-6.1-terra`, `grok-4.5`, `grok-4.6`, `grok-4.7`, `qwen3.8-max-cjx`, `xiaozhanAI`
- **Gemini Native** (`/v1beta/...`) — 6 models: `gemini-3.1-pro-low`, `gemini-3.1-pro-preview-1020`, `gemini-3.8-flash`, `gemini-3.8-flash-nx`, `gemini-4-argon`, `gemini-4-argon-nx`

For a model × protocol combination with no reachable channel, the gateway returns `503 no_channel_available` (the model exists but has no channel on that protocol — **not a 404**); switch to a supported protocol.

## Legacy ids (aliases)

135 older ids still resolve onto the canonical entries above, so integrations pinned to them keep working. An id that is neither in the tables above nor an alias returns `model_not_found` — the catalog endpoint is the source of truth.

## Live data

The model list and prices track upstream changes. For live data see [https://qd.zsopc.com/models](https://qd.zsopc.com/models) or the public catalog endpoint (no auth required):

```bash
curl https://qd.zsopc.com/api/v1/models/public | jq .
```

The response looks like `{ "models": [...] }`. Note that this endpoint returns **every registered entry** — including placeholder SKUs that aren't live yet and non-chat SKUs — so filter on the three fields below when consuming it from a script rather than looping over the whole list:

| Field | Meaning |
|---|---|
| `callable` | `false` means a placeholder SKU (listed but not live); calling it returns `model_not_found` |
| `endpoint_type` | `null` means an ordinary chat model. `images_generations`, `videos_generations`, `contents_generations_tasks` and similar mean the model uses its own dedicated endpoint and cannot be sent to `/v1/chat/completions` — see [Image / video / music APIs](./media-apis.md) |
| `supported_protocols` | the client protocols available for this model (see above) |

To take only the models you can send straight to `/v1/chat/completions`:

```bash
curl -s https://qd.zsopc.com/api/v1/models/public \
  | jq -r '.models[] | select(.callable and (.supported_protocols | index("openai_chat"))) | .id'
```
