This document provides a detailed summary of the available API endpoints, including explanations, a complete list of supported parameters, and `curl` examples.
Note: Replace your_api_key with your actual API key and MODEL_NAME with the name of the model you want to use.
Base path: /v1
| Parameter | Type | Required | Description |
|---|---|---|---|
| model | string | required | The chosen model to use for generation. |
| multi_models | array of strings | optional | List of models to use for generation. Chosen randomly during the request. |
| multiplier | number | optional | Multiplier for a finetuned model's LoRA alpha value. Higher value means stronger effect from the finetune. 0.5x, 2x and 4x supported. |
| n | integer | optional | Number of output sequences to return for the given prompt. |
| min_tokens | integer | optional | Minimum number of tokens to generate per output sequence before EOS or stop tokens are generated. |
| presence_penalty | number | optional | Float that penalizes new tokens based on whether they appear in the generated text so far. Values > 0 encourage the model to use new tokens, while values < 0 encourage the model to repeat tokens. |
| frequency_penalty | number | optional | Float that penalizes new tokens based on their frequency in the generated text so far. Values > 0 encourage the model to use new tokens, while values < 0 encourage the model to repeat tokens. |
| repetition_penalty | number | optional | Float that penalizes new tokens based on their frequency in the generated text so far. Freq_pen is applied additively while rep_pen is applied multiplicatively. Must be in [1, inf). Set to 1 to disable the effect. |
| no_repeat_ngram_size | integer | optional | (Aphrodite only) Size of the n-grams to prevent repeating. 1 would mean no token can appear twice. 2 would mean no pair of consecutive tokens can appear twice. |
| temperature | number | optional | Float that controls the randomness of the sampling. Lower values make the model more deterministic, while higher values make the model more random. Zero means greedy sampling. |
| top_p | number | optional | Float that controls the cumulative probability of the top tokens to consider. Must be in (0, 1]. Set to 1 to consider all tokens. |
| top_k | integer | optional | Integer that controls the number of top tokens to consider. Set to -1 to consider all tokens. |
| top_a | number | optional | (Aphrodite only) Float that controls the cutoff for Top-A sampling. Exact cutoff is top_a*max_prob**2. Must be in [0, inf], 0 to disable. |
| min_p | number | optional | Float that controls the cutoff for min-p sampling. Exact cutoff is min_p*max_prob. Must be in [0, 1], 0 to disable. |
| tfs | number | optional | (Aphrodite only) Float that controls the cumulative approximate curvature of the distribution to retain for Tail Free Sampling. Must be in (0, 1]. Set to 1 to disable. |
| eta_cutoff | number | optional | (Aphrodite only) Float that controls the cutoff threshold for Eta sampling (a form of entropy adaptive truncation sampling). Threshold is computed as min(eta, sqrt(eta)*entropy(probs)). Specified in units of 1e-4. Set to 0 to disable. |
| epsilon_cutoff | number | optional | (Aphrodite only) Float that controls the cutoff threshold for Epsilon sampling (simple probability threshold truncation). Specified in units of 1e-4. Set to 0 to disable. |
| typical_p | number | optional | (Aphrodite only) Float that controls the cumulative probability of tokens closest in surprise to the expected surprise. Must be in (0, 1]. Set to 1 to disable. |
| dynatemp_min | number | optional | (Aphrodite only) Minimum temperature for dynamic temperature sampling. Range [0, inf). |
| dynatemp_max | number | optional | (Aphrodite only) Maximum temperature for dynamic temperature sampling. Range [0, inf). |
| dynatemp_exponent | number | optional | (Aphrodite only) Exponent for dynamic temperature sampling. Range [0, inf). |
| smoothing_factor | number | optional | (Aphrodite only) Smoothing factor for Quadratic Sampling. |
| smoothing_curve | number | optional | (Aphrodite only) Smoothing curve for Cubic Sampling. |
| seed | integer | optional | Random seed to use for the generation. |
| length_penalty | number | optional | (Aphrodite only) Penalizes sequences based on their length. Used in beam search. |
| stop | string or array of strings | optional | List of strings that stop the generation when they are generated. The returned output will not contain the stop strings. |
| stop_token_ids | array of integers | optional | List of token IDs that stop the generation when they are generated. The returned output will contain the stop tokens unless they are special tokens. |
| include_stop_str_in_output | boolean | optional | Whether to include the stop strings in the output text. Defaults to False. |
| ignore_eos | boolean | optional | Whether to ignore the EOS token and continue generating tokens after the EOS token is generated. |
| max_tokens | integer | optional | Maximum number of tokens to generate per output sequence. |
| logprobs | integer | optional | Number of log probabilities to return per output token. When set to None, no probability is returned. |
| prompt_logprobs | integer | optional | Number of log probabilities to return per prompt token. |
| detokenize | boolean | optional | Whether to detokenize the output. Defaults to True. |
| skip_special_tokens | boolean | optional | Whether to skip special tokens in the output. Defaults to True. |
| spaces_between_special_tokens | boolean | optional | Whether to add spaces between special tokens in the output. Defaults to True. |
| logits_processors | array | optional | List of functions that modify logits based on previously generated tokens and optionally prompt tokens. |
| logit_bias | object | optional | List of LogitsProcessors to change the probability of token prediction at runtime. |
| truncate_prompt_tokens | integer | optional | If set to an integer k, will use only the last k tokens from the prompt (left-truncation). Default: None (no truncation). |
| xtc_threshold | number | optional | (Aphrodite only) In XTC sampling, if 2 or more tokens have a probability above this threshold, consider removing all but the last one. Disabled: 0. |
| xtc_probability | number | optional | (Aphrodite only) The probability that the removal will happen in XTC sampling. Set to 0 to disable. Default: 0. |
| guided_json | string or object | optional | If specified, the output will follow the JSON schema. Can be a JSON string or a Python dictionary. |
| guided_regex | string | optional | If specified, the output will follow the regex pattern. |
| guided_choice | array of strings | optional | If specified, the output will be exactly one of the provided choices (a list of strings). |
| guided_decoding_backend | string | optional | Overrides the default guided decoding backend for this specific request. Must be either "outlines" or "lm-format-enforcer". |
| guided_whitespace_pattern | string | optional | Overrides the default whitespace pattern for guided JSON decoding. |
| nsigma | number | optional | (Aphrodite only) Number of standard deviations from the maximum logit to use as a cutoff threshold. Tokens with logits below (max_logit - nsigma * std_dev) are filtered out. Higher values (e.g. 3.0) keep more tokens, lower values (e.g. 1.0) are more selective. Must be positive. 0 to disable. |
| dry_multiplier | number | optional | (Aphrodite only) Float that controls the magnitude of the DRY sampling penalty. Higher values create stronger penalties against repetition. The penalty is multiplied by this value before being applied. Must be non-negative. 0 disables the sampler. |
| dry_base | number | optional | (Aphrodite only) Base for the exponential growth of the DRY sampling penalty. Controls how quickly the penalty increases with longer repeated sequences. Must be greater than 1. Higher values (e.g. 2.0) create more aggressive penalties for longer repetitions. Defaults to 1.75. |
| dry_allowed_length | integer | optional | (Aphrodite only) Maximum number of tokens that can be repeated without incurring a DRY sampling penalty. Sequences longer than this will be penalized exponentially. Must be at least 1. Defaults to 2. |
| dry_sequence_breaker_ids | array of integers | optional | (Aphrodite only) List of token IDs that stop the matching of repeated content. These tokens will break up the input into sections where repetition is evaluated separately. Common examples are newlines, quotes, and other structural tokens. Defaults to None. |
| dry_range | integer | optional | (Aphrodite only) The range of tokens (input + output) to apply the DRY sampler. |
| skew | number | optional | (Aphrodite only) Bias the token selection towards higher or lower probability tokens. Defaults to 0 (disabled). |
| sampler_priority | array | optional | (Aphrodite only) A list of integers to control the order in which samplers are applied. |
| allowed_token_ids | array of integers | optional | If provided, the engine will construct a logits processor which only retains scores for the given token ids. |
| bad_words | array of strings | optional | List of words that are not allowed to be generated. |
| banned_phrases_token_ids | array of array of integers | optional | (Aphrodite only) List of token sequences that are not allowed to be generated. |
| best_of | integer | optional | Number of output sequences that are generated from the prompt. From these `best_of` sequences, the top `n` sequences are returned. `best_of` must be greater than or equal to `n`. By default, `best_of` is set to `n`. |
| custom_token_bans | array of integers | optional | (Aphrodite only) List of token IDs to ban from generating. |
| deepconf_threshold | number | optional | (Aphrodite only) Confidence threshold for early stopping in DeepConf. |
| deepconf_window_size | integer | optional | (Aphrodite only) Size of the sliding window for confidence calculation in DeepConf. |
| dry_early_exit_match_len | integer | optional | (Aphrodite only) If we find this large a match in DRY sampling, we stop searching. |
| dry_max_ngram | integer | optional | (Aphrodite only) Maximum length of match to check in DRY sampling. |
| dry_max_occurrences | integer | optional | (Aphrodite only) How many occurrences of last_token we analyze in DRY sampling. |
| early_stopping | boolean or string | optional | (Aphrodite only) Controls the stopping condition for beam search. It accepts `True`, `False`, or `"never"`. |
| enable_deepconf | boolean | optional | (Aphrodite only) Enable DeepConf for confidence-based early stopping. |
| extra_args | object | optional | Extra arguments to pass to the engine. |
| guided_decoding | object | optional | If provided, the engine will construct a guided decoding logits processor from these parameters. |
| mirostat_eta | number | optional | (Aphrodite only) Rate at which mirostat updates its internal surprisal value. Range [0, inf). |
| mirostat_mode | integer | optional | (Aphrodite only) Can either be 0 (disabled) or 2 (Mirostat v2). |
| mirostat_tau | number | optional | (Aphrodite only) Target "surprisal" that mirostat works towards. Range [0, inf). |
| output_kind | string | optional | The type of output to generate. Can be "cumulative", "delta", or "final_only". |
| structured_outputs | object | optional | (vLLM only) Parameters for configuring structured outputs. |
| temperature_last | boolean | optional | (Aphrodite only) Whether to use temperature as the last sampler in the sampling pipeline. |
| token_ban_ranges | array | optional | (Aphrodite only) List of tuples (tokens, start, length) to ban from generating. |
| use_beam_search | boolean | optional | (Aphrodite only) Whether to use beam search instead of sampling. |
Handles chat completions. This endpoint takes a list of messages and returns a generated response.
Handles text completions. This endpoint takes a prompt and returns a generated response.
Tokenizes the given text.
| Parameter | Type | Required | Description |
|---|---|---|---|
| model | string | optional | The model to use for tokenization. |
| multi_models | array of strings | optional | A list of models to choose from. |
| prompt | string | optional | The prompt to tokenize. |
| messages | array | optional | The messages to tokenize. |
Handles image-to-image generation.
| Parameter | Type | Required | Description |
|---|---|---|---|
| sd_model_checkpoint | string | required | The name of the model checkpoint. |
| prompt | string | required | The text prompt. |
| init_images | array of strings | required | Base64-encoded initial images. |
| negative_prompt | string | optional | The negative prompt. |
| steps | integer | optional | Number of sampling steps. |
| sampler_name | string | optional | Sampling method. |
| width | integer | optional | Image width. |
| height | integer | optional | Image height. |
| clip_skip | integer | optional | Number of CLIP layers to skip. |
| seed | integer | optional | Random seed. |
| cfg_scale | number | optional | Classifier-Free Guidance scale. |
| stream | boolean | optional | Whether to stream the response. |
| batch_size | integer | optional | Number of images to generate in a batch. |
| denoising_strength | number | optional | Denoising strength for img2img. |
| mask | string | optional | Base64-encoded mask for inpainting. |
| mask_blur | integer | optional | Mask blur for inpainting. |
| inpainting_fill | integer | optional | Inpainting fill mode. |
| inpaint_full_res | boolean | optional | Whether to inpaint at full resolution. |
| inpaint_full_res_padding | integer | optional | Padding for full-resolution inpainting. |
| inpainting_mask_invert | integer | optional | Whether to invert the inpainting mask. |
| initial_noise_multiplier | number | optional | Initial noise multiplier. |
| detailer_enabled | boolean | optional | Enable the detailer. |
| detailer_prompt | string | optional | Prompt for the detailer. |
| detailer_negative | string | optional | Negative prompt for the detailer. |
| detailer_steps | integer | optional | Steps for the detailer. |
| detailer_strength | number | optional | Strength of the detailer. |
| detailer_model | string | optional | Model for the detailer. |
| detailer_classes | string | optional | Classes for the detailer. |
| detailer_conf | number | optional | Confidence for the detailer. |
| detailer_max | integer | optional | Max detections for the detailer. |
| detailer_iou | number | optional | IoU for the detailer. |
| detailer_padding | integer | optional | Padding for the detailer. |
| detailer_blur | integer | optional | Blur for the detailer. |
| detailer_merge | boolean | optional | Merge mode for the detailer. |
| schedulers_rescale_betas | boolean | optional | Rescale betas for schedulers. |
| schedulers_use_thresholding | boolean | optional | Use thresholding for schedulers. |
| schedulers_sigma | number | optional | Sigma for schedulers. |
| schedulers_beta_schedule | string | optional | Beta schedule for schedulers. |
| scheduler_eta | number | optional | ETA for schedulers. |
| schedulers_solver_order | integer | optional | Solver order for schedulers. |
| schedulers_beta_start | number | optional | Beta start for schedulers. |
| schedulers_beta_end | number | optional | Beta end for schedulers. |
| schedulers_timesteps_range | string | optional | Timesteps range for schedulers. |
| schedulers_shift | number | optional | Shift for schedulers. |
| schedulers_sigma_adjust | boolean | optional | Sigma adjustment for schedulers. |
| schedulers_sigma_adjust_min | number | optional | Min sigma adjustment. |
| schedulers_sigma_adjust_max | number | optional | Max sigma adjustment. |
Handles text-to-image generation.
| Parameter | Type | Required | Description |
|---|---|---|---|
| sd_model_checkpoint | string | required | The name of the model checkpoint. |
| prompt | string | required | The text prompt. |
| negative_prompt | string | optional | The negative prompt. |
| steps | integer | optional | Number of sampling steps. |
| sampler_name | string | optional | Sampling method. |
| width | integer | optional | Image width. |
| height | integer | optional | Image height. |
| clip_skip | integer | optional | Number of CLIP layers to skip. |
| seed | integer | optional | Random seed. |
| cfg_scale | number | optional | Classifier-Free Guidance scale. |
| stream | boolean | optional | Whether to stream the response. |
| batch_size | integer | optional | Number of images to generate in a batch. |
| hr_sampler_name | string | optional | Sampler name for high-res fix. |
| detailer_enabled | boolean | optional | Enable the detailer. |
| detailer_prompt | string | optional | Prompt for the detailer. |
| detailer_negative | string | optional | Negative prompt for the detailer. |
| detailer_steps | integer | optional | Steps for the detailer. |
| detailer_strength | number | optional | Strength of the detailer. |
| detailer_model | string | optional | Model for the detailer. |
| detailer_classes | string | optional | Classes for the detailer. |
| detailer_conf | number | optional | Confidence for the detailer. |
| detailer_max | integer | optional | Max detections for the detailer. |
| detailer_iou | number | optional | IoU for the detailer. |
| detailer_padding | integer | optional | Padding for the detailer. |
| detailer_blur | integer | optional | Blur for the detailer. |
| detailer_merge | boolean | optional | Merge mode for the detailer. |
| schedulers_rescale_betas | boolean | optional | Rescale betas for schedulers. |
| schedulers_use_thresholding | boolean | optional | Use thresholding for schedulers. |
| schedulers_sigma | number | optional | Sigma for schedulers. |
| schedulers_beta_schedule | string | optional | Beta schedule for schedulers. |
| scheduler_eta | number | optional | ETA for schedulers. |
| schedulers_solver_order | integer | optional | Solver order for schedulers. |
| schedulers_beta_start | number | optional | Beta start for schedulers. |
| schedulers_beta_end | number | optional | Beta end for schedulers. |
| schedulers_timesteps_range | string | optional | Timesteps range for schedulers. |
| schedulers_shift | number | optional | Shift for schedulers. |
| schedulers_sigma_adjust | boolean | optional | Sigma adjustment for schedulers. |
| schedulers_sigma_adjust_min | number | optional | Min sigma adjustment. |
| schedulers_sigma_adjust_max | number | optional | Max sigma adjustment. |
Upscales a single image.
| Parameter | Type | Required | Description |
|---|---|---|---|
| image | string | required | The base64-encoded image to upscale. |
| upscaler_1 | string | optional | The name of the upscaler to use. |
| resize_mode | integer | optional | The resize mode. |
| upscaling_resize | number | optional | The factor by which to resize the image. |
Retrieves available text generation models. No parameters.
Retrieves available image generation models. No parameters.
Retrieves available upscalers. No parameters.
Retrieves image generation options. No parameters.
Retrieves available image samplers. No parameters.
Retrieves parallel request limits. No parameters.
Base path: /sdapi/v1
The endpoints under /sdapi/v1 are designed for compatibility with the Stable Diffusion API and share the same parameters as their /v1 counterparts.
See /v1/img2img for parameters.
See /v1/txt2img for parameters.
See /v1/upscale-img for parameters.
Retrieves available image models.
Retrieves available upscalers.
Retrieves image generation options.
Retrieves available image samplers.