Task
Chat models
Chat and reasoning models on OpenAI-compatible /v1/chat/completions, with streaming. Per-model token rates, billed to the same balance as media models.
- Models
- 26
- From
- $0.425 per 1 million tokens
- To
- $45 per 1 million tokens
- Outputs
- Text
Where to start
A few families worth opening first. Every model and price in this category is listed in the table below.
Chat models26 models
- Grok 4.7
xai/grok-4.7/chat
- Price
- Input $2Output $6/ 1 million tokens
- Best for
- Chat
- Content limits
- Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer.
- DeepSeek V4.1 Flash
deepseek/v4.1-flash/chat
- Price
- Input $0.30Output $1.20/ 1 million tokens
- Best for
- Chat
- Content limits
- Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer.
- Grok 4.3
xai/grok-4.3/chat
- Price
- Input $1.25Output $2.50/ 1 million tokens
- Best for
- Chat
- Content limits
- Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer.
- Claude Haiku 5.5
anthropic/claude-haiku-5.5/chat
- Price
- Input $0.10Output $0.50/ 1 million tokens
- Best for
- Chat
- Content limits
- Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer.
- Claude Sonnet 5.5
anthropic/claude-sonnet-5.5/chat
- Price
- Input $2Output $10/ 1 million tokens
- Best for
- Chat
- Content limits
- Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer.
- Claude Opus 5.5
anthropic/claude-opus-5.5/chat
- Price
- Input Anthropic official price: $4 $3.60Output Anthropic official price: $20 $18/ 1 million tokens10% below the publisher's list price
- Best for
- Chat
- Content limits
- Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer.
- GLM 5.3 Flash
zai/glm-5.3-flash/chat
- Price
- Input $0.1275Output $0.425/ 1 million tokens
- Best for
- Chat
- Content limits
- Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer.
- GLM 5.3
zai/glm-5.3/chat
- Price
- Input Z.ai official price: $1.40 $1.05Output Z.ai official price: $4.40 $3.30/ 1 million tokens25% below the publisher's list price
- Best for
- Chat
- Content limits
- Light — We add no content filter, but this model goes past its own limits only some of the time, even with a fiction-writing system prompt. Expect refusals and plan for retries. Legal and consent duties remain with the customer.
- DeepSeek V4 Pro
deepseek/v4-pro/chat
- Price
- Input $1.32Output $3.96/ 1 million tokens
- Best for
- Chat
- Content limits
- Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer.
- Gemini 3.7 Flash
google/gemini-3.7-flash/chat
- Price
- Input Google official price: $0.75 $0.675Output Google official price: $3.75 $3.375/ 1 million tokens10% below the publisher's list price
- Best for
- Chat
- Content limits
- Light — We add no content filter, but this model goes past its own limits only some of the time, even with a fiction-writing system prompt. Expect refusals and plan for retries. Legal and consent duties remain with the customer.
- Grok 4.6
xai/grok-4.6/chat
- Price
- Input $2Output $6/ 1 million tokens
- Best for
- Chat
- Content limits
- Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer.
- DeepSeek V4 Flash
deepseek/v4-flash/chat
- Price
- Input $0.44Output $1.32/ 1 million tokens
- Best for
- Chat
- Content limits
- Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer.
- Claude Opus 5
anthropic/claude-opus-5/chat
- Price
- Input Anthropic official price: $5 $4.50Output Anthropic official price: $25 $22.50/ 1 million tokens10% below the publisher's list price
- Best for
- Chat
- Content limits
- Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer.
- Gemini 3.5 Flash-Lite
google/gemini-3.5-flash-lite/chat
- Price
- Input $0.30Output $2.50/ 1 million tokens
- Best for
- Chat
- Content limits
- Light — We add no content filter, but this model goes past its own limits only some of the time, even with a fiction-writing system prompt. Expect refusals and plan for retries. Legal and consent duties remain with the customer.
- Kimi K3
moonshot/kimi-k3/chat
- Price
- Input Moonshot AI official price: $3 $2.70Output Moonshot AI official price: $15 $13.50/ 1 million tokens10% below the publisher's list price
- Best for
- Chat
- Content limits
- Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer.
- Grok 4.5
xai/grok-4.5/chat
- Price
- Input $2Output $6/ 1 million tokens
- Best for
- Chat
- Content limits
- Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer.
- Claude Sonnet 5
anthropic/claude-sonnet-5/chat
- Price
- Input Anthropic official price: $2 $1.90Output Anthropic official price: $10 $9.50/ 1 million tokens5% below the publisher's list price
- Best for
- Chat
- Content limits
- Filtered — This model refuses requests beyond its own limits, including when your system prompt sets up a fiction-writing context. Measured, not assumed — models we have not measured carry no label at all.
- GLM 5.2
zai/glm-5.2/chat
- Price
- Input Z.ai official price: $1.40 $0.98Output Z.ai official price: $4.40 $3.08/ 1 million tokens30% below the publisher's list price
- Best for
- Chat
- Content limits
- Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer.
- Claude Fable 5
anthropic/claude-fable-5/chat
- Price
- Input Anthropic official price: $10 $9Output Anthropic official price: $50 $45/ 1 million tokens10% below the publisher's list price
- Best for
- Chat
- Content limits
- Softened — This model answers instead of refusing, but requests beyond its own limits come back toned down, even with a fiction-writing system prompt. The call still succeeds and is still billed, so read the reply rather than trusting the status code.
- Gemini 3.1 Pro Preview
google/gemini-3.1-pro-preview/chat
- Price
- Input Google official price: $2 $1.90Output Google official price: $12 $11.40/ 1 million tokens5% below the publisher's list price
- Best for
- Chat
- Content limits
- Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer.
- Claude Sonnet 4.6
anthropic/claude-sonnet-4.6/chat
- Price
- Input $3Output $15/ 1 million tokens
- Best for
- Chat
- Content limits
- Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer.
- Claude Opus 4.6
anthropic/claude-opus-4.6/chat
- Price
- Input $5Output $25/ 1 million tokens
- Best for
- Chat
- Content limits
- Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer.
- Gemini 3 Flash Preview
google/gemini-3-flash-preview/chat
- Price
- Input Google official price: $0.50 $0.45Output Google official price: $3 $2.70/ 1 million tokens10% below the publisher's list price
- Best for
- Chat
- Content limits
- Light — We add no content filter, but this model goes past its own limits only some of the time, even with a fiction-writing system prompt. Expect refusals and plan for retries. Legal and consent duties remain with the customer.
- Claude Opus 4.5
anthropic/claude-opus-4.5/chat
- Price
- Input Anthropic official price: $5 $4.50Output Anthropic official price: $25 $22.50/ 1 million tokens10% below the publisher's list price
- Best for
- Chat
- Content limits
- Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer.
- Claude Sonnet 4.5
anthropic/claude-sonnet-4.5/chat
- Price
- Input Anthropic official price: $3 $2.70Output Anthropic official price: $15 $13.50/ 1 million tokens10% below the publisher's list price
- Best for
- Chat
- Content limits
- Softened — This model answers instead of refusing, but requests beyond its own limits come back toned down, even with a fiction-writing system prompt. The call still succeeds and is still billed, so read the reply rather than trusting the status code.
- Gemini 2.5 Flash
google/gemini-2.5-flash/chat
- Price
- Input Google official price: $0.30 $0.285Output Google official price: $2.50 $2.375/ 1 million tokens5% below the publisher's list price
- Best for
- Chat
- Content limits
- Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer.
| Model | Best for | Content limits | View details | |
|---|---|---|---|---|
| Grok 4.7xai/grok-4.7/chat | Chat | Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer. | Input $2Output $6/ 1 million tokens | |
| DeepSeek V4.1 Flashdeepseek/v4.1-flash/chat | Chat | Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer. | Input $0.30Output $1.20/ 1 million tokens | |
| Grok 4.3xai/grok-4.3/chat | Chat | Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer. | Input $1.25Output $2.50/ 1 million tokens | |
| Claude Haiku 5.5anthropic/claude-haiku-5.5/chat | Chat | Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer. | Input $0.10Output $0.50/ 1 million tokens | |
| Claude Sonnet 5.5anthropic/claude-sonnet-5.5/chat | Chat | Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer. | Input $2Output $10/ 1 million tokens | |
| Claude Opus 5.5anthropic/claude-opus-5.5/chat | Chat | Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer. | Input Anthropic official price: $4 $3.60Output Anthropic official price: $20 $18/ 1 million tokens10% below the publisher's list price | |
| GLM 5.3 Flashzai/glm-5.3-flash/chat | Chat | Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer. | Input $0.1275Output $0.425/ 1 million tokens | |
| GLM 5.3zai/glm-5.3/chat | Chat | Light — We add no content filter, but this model goes past its own limits only some of the time, even with a fiction-writing system prompt. Expect refusals and plan for retries. Legal and consent duties remain with the customer. | Input Z.ai official price: $1.40 $1.05Output Z.ai official price: $4.40 $3.30/ 1 million tokens25% below the publisher's list price | |
| DeepSeek V4 Prodeepseek/v4-pro/chat | Chat | Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer. | Input $1.32Output $3.96/ 1 million tokens | |
| Gemini 3.7 Flashgoogle/gemini-3.7-flash/chat | Chat | Light — We add no content filter, but this model goes past its own limits only some of the time, even with a fiction-writing system prompt. Expect refusals and plan for retries. Legal and consent duties remain with the customer. | Input Google official price: $0.75 $0.675Output Google official price: $3.75 $3.375/ 1 million tokens10% below the publisher's list price | |
| Grok 4.6xai/grok-4.6/chat | Chat | Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer. | Input $2Output $6/ 1 million tokens | |
| DeepSeek V4 Flashdeepseek/v4-flash/chat | Chat | Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer. | Input $0.44Output $1.32/ 1 million tokens | |
| Claude Opus 5anthropic/claude-opus-5/chat | Chat | Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer. | Input Anthropic official price: $5 $4.50Output Anthropic official price: $25 $22.50/ 1 million tokens10% below the publisher's list price | |
| Gemini 3.5 Flash-Litegoogle/gemini-3.5-flash-lite/chat | Chat | Light — We add no content filter, but this model goes past its own limits only some of the time, even with a fiction-writing system prompt. Expect refusals and plan for retries. Legal and consent duties remain with the customer. | Input $0.30Output $2.50/ 1 million tokens | |
| Kimi K3moonshot/kimi-k3/chat | Chat | Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer. | Input Moonshot AI official price: $3 $2.70Output Moonshot AI official price: $15 $13.50/ 1 million tokens10% below the publisher's list price | |
| Grok 4.5xai/grok-4.5/chat | Chat | Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer. | Input $2Output $6/ 1 million tokens | |
| Claude Sonnet 5anthropic/claude-sonnet-5/chat | Chat | Filtered — This model refuses requests beyond its own limits, including when your system prompt sets up a fiction-writing context. Measured, not assumed — models we have not measured carry no label at all. | Input Anthropic official price: $2 $1.90Output Anthropic official price: $10 $9.50/ 1 million tokens5% below the publisher's list price | |
| GLM 5.2zai/glm-5.2/chat | Chat | Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer. | Input Z.ai official price: $1.40 $0.98Output Z.ai official price: $4.40 $3.08/ 1 million tokens30% below the publisher's list price | |
| Claude Fable 5anthropic/claude-fable-5/chat | Chat | Softened — This model answers instead of refusing, but requests beyond its own limits come back toned down, even with a fiction-writing system prompt. The call still succeeds and is still billed, so read the reply rather than trusting the status code. | Input Anthropic official price: $10 $9Output Anthropic official price: $50 $45/ 1 million tokens10% below the publisher's list price | |
| Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview/chat | Chat | Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer. | Input Google official price: $2 $1.90Output Google official price: $12 $11.40/ 1 million tokens5% below the publisher's list price | |
| Claude Sonnet 4.6anthropic/claude-sonnet-4.6/chat | Chat | Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer. | Input $3Output $15/ 1 million tokens | |
| Claude Opus 4.6anthropic/claude-opus-4.6/chat | Chat | Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer. | Input $5Output $25/ 1 million tokens | |
| Gemini 3 Flash Previewgoogle/gemini-3-flash-preview/chat | Chat | Light — We add no content filter, but this model goes past its own limits only some of the time, even with a fiction-writing system prompt. Expect refusals and plan for retries. Legal and consent duties remain with the customer. | Input Google official price: $0.50 $0.45Output Google official price: $3 $2.70/ 1 million tokens10% below the publisher's list price | |
| Claude Opus 4.5anthropic/claude-opus-4.5/chat | Chat | Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer. | Input Anthropic official price: $5 $4.50Output Anthropic official price: $25 $22.50/ 1 million tokens10% below the publisher's list price | |
| Claude Sonnet 4.5anthropic/claude-sonnet-4.5/chat | Chat | Softened — This model answers instead of refusing, but requests beyond its own limits come back toned down, even with a fiction-writing system prompt. The call still succeeds and is still billed, so read the reply rather than trusting the status code. | Input Anthropic official price: $3 $2.70Output Anthropic official price: $15 $13.50/ 1 million tokens10% below the publisher's list price | |
| Gemini 2.5 Flashgoogle/gemini-2.5-flash/chat | Chat | Uncensored — We add no content filter to this model, but like most chat models it declines a blunt request that comes with no context. It writes unrestricted fiction reliably once your system prompt sets up a fiction-writing context, or once the request is framed as a story. Legal and consent duties remain with the customer. | Input Google official price: $0.30 $0.285Output Google official price: $2.50 $2.375/ 1 million tokens5% below the publisher's list price |
- Uncensored
- No filter on our side — set a fiction-writing system prompt; blunt asks with no context get declined.
- Light
- Goes past its own limits only some of the time, even with a fiction-writing system prompt.
- Softened
- Still answers and still bills — requests beyond its limits come back toned down.
- Filtered
- Refuses requests beyond its own limits, even with a fiction-writing system prompt.
Use cases
Companion and roleplay backends
The text half of a companion product: persona, system prompt, memory, dialogue. Nothing is rewritten on the way through — the system prompt reaches the model exactly as you wrote it, and there is no per-request declaration to fill in. What varies is each model’s own alignment: some hold a second-person scene for hundreds of turns, others slide back into assistant voice halfway. Run one real conversation through two of them and keep whichever stays in character.
Prompt expansion for the media models
Use an LLM to turn a user’s description into a clear shot prompt, then submit the media task. Both calls share an API key and balance for simpler management.
Agents and long-document work
Compare context capacity, response speed and rates for long documents or multi-step tasks. Choose a model for each step and pay for actual usage.
Frequently asked questions
Yes. Text models are served at /v1/chat/completions with the request and response shapes the OpenAI SDKs expect, streaming included. In most codebases the migration is a base URL, an API key and a model name — the rest of your client code is untouched.
The platform applies no classifier and requires no content declaration, so the prompt reaches the model as written. What still applies is each model’s own alignment, and it differs noticeably from one to the next. Rather than trusting any published ranking, run the same conversation through two and keep the one that holds the character.
That is what most of the text traffic here is doing. Because nothing sits between your system prompt and the model, the persona is yours to define at whatever length you need, and long-context models carry the character sheet alongside everything said so far. Holding a voice across a long conversation and writing unrestricted prose are two different properties, though — a model that does the first well will not necessarily do the second. Test with the system prompt you actually ship, not a sample one.
Token usage is billed in US dollars at the selected model’s rates. See its pricing details for input, output and any other usage categories. Check the displayed unit rather than assuming every rate is per thousand tokens.
Yes. Chat, image and video models share an account, API key and balance, making it easier to combine prompt development and media generation in one workflow.
About this category
Explore chat completions models from the current catalogue. Check each model page for input fields, parameter limits, pricing and availability. SpicyAPI adds no content classifier; capabilities depend on the selected model. Text uses an OpenAI-compatible interface and shares your account and balance with image and video calls. Failed or timed-out tasks are usually not charged. Some models charge for content refusals as stated on their model page; check those billing terms before submitting.