GET

/v1/schemas/fireworks/chat/v1%2Fchat%2Fcompletions

kind inputactivity chathash 3acf8ebe…extracted Oct 6, 2026version currentsource upstream spec ↗

Request body

49 properties · 1 required · 18 $defs
propertytypedescriptionconstraints
model*stringThe name of the model to use.
messagesChatMessage[]A list of messages comprising the conversation so far. When ``prompt_token_ids`` is supplied this field is ignored (the caller is taking responsibility for chat-template rendering and tokenization upstream).
toolsChatCompletionTool[]A list of tools the model may call. Currently, only functions are supported as a tool.
tool_choiceenum | FunctionSelection | AllowedToolsS…Controls which (if any) tool is called by the model.default: "auto"
streambooleanWhether to stream back partial progress. If set, tokens will be sent as data-only [server-sent events](https://developer.mozilla.org/en-US/docs/Web/API/Server-sent_events/Using_server-sent_events#Event_stream_format) as they become available, with the stream terminated by a `data: [DONE]` message.nullable
response_formatResponseFormatAllows to force the model to produce specific output format. Setting to `{ "type": "json_object" }` enables JSON mode, which guarantees the message the model generates is valid JSON. If `"type"` is `"json_schema"`, a JSON schema must be provided. E.g., `response_format = {"type": "json_schema", "json_schema": <json_schema>}`. Important: when using JSON mode, it's crucial to also instruct the model to produce JSON via a system or user message. Without this, the model may generate an unending stream of whitespace until the generation reaches the token limit, resulting in a long-running and seemingly "stuck" request. Also note that the message content may be partially cut off if `finish_reason="length"`, which indicates the generation exceeded `max_tokens` or the conversation exceeded the max context length. In this case the return value might not be a valid JSON.nullable
temperaturenumberWhat sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both. Required range: `0 <= x <= 2` Example: `1`nullable
top_kintegerTop-k sampling is another sampling method where the k most probable next tokens are filtered and the probability mass is redistributed among only those k next tokens. The value of k controls the number of candidates for the next token at each step during text generation. Must be between 0 and 100. Required range: `0 <= x <= 100` Example: `50`nullable
userstringA unique identifier representing your end-user, which can help monitor and detect abuse.nullable
prompt_cache_keystringA key used for prompt caching session affinity. Requests with the same prompt_cache_key are routed to the same backend to maximize KV cache hit rates. This is the preferred field for session affinity (takes priority over the 'user' field).nullable
prompt_cache_isolation_keystringIsolation key for prompt caching to separate cache entries.nullable
raw_outputbooleanReturn raw output from the model.nullable
perf_metrics_in_responsebooleanWhether to include performance metrics in the response body. **Non-streaming requests:** Performance metrics are always included in response headers (e.g., `fireworks-prompt-tokens`, `fireworks-server-time-to-first-token`). Setting this to `true` additionally includes the same metrics in the response body under the `perf_metrics` field. **Streaming requests:** Performance metrics are only included in the response body under the `perf_metrics` field in the final chunk (when `finish_reason` is set). This is because headers may not be accessible during streaming. The response body `perf_metrics` field contains the following metrics: **Basic Metrics (all deployments):** - `prompt-tokens`: Number of tokens in the prompt - `cached-prompt-tokens`: Number of cached prompt tokens - `server-time-to-first-token`: Time from request start to first token (in seconds) - `server-processing-time`: Total processing time (in seconds, only for completed requests) **Predicted Outputs Metrics:** - `speculation-prompt-tokens`: Number of speculative prompt tokens - `speculation-prompt-matched-tokens`: Number of matched speculative prompt tokens (for completed requests) **Dedicated Deployment Only Metrics:** - `speculation-generated-tokens`: Number of speculative generated tokens (for completed requests) - `speculation-acceptance`: Speculation acceptance rates by position - `backend-host`: Hostname of the backend server - `num-concurrent-requests`: Number of concurrent requests - `deployment`: Deployment name - `tokenizer-queue-duration`: Time spent in tokenizer queue - `tokenizer-duration`: Time spent in tokenizer - `prefill-queue-duration`: Time spent in prefill queue - `prefill-duration`: Time spent in prefill - `generation-queue-duration`: Time spent in generation queue - `generation-duration`: Time spent in generationnullable
stream_optionsStreamOptionsOptions for streaming responses. Only valid with ``stream=true``. Fireworks includes a final SSE chunk carrying usage totals by default; set ``include_usage=false`` to opt out. The ``buffer_tokens`` / ``buffer_ms`` / ``buffer_mode`` fields coalesce SSE chunks by token count and/or time (vLLM backend; override the deployment default).nullable
nintegerHow many completions to generate for each prompt.default: 1
service_tierenumThe service tier to use for the request. Specifies the processing type used for serving the request. Only "priority" is supported, while all other values will be treated as "default" tier.auto | default | flex | priority · default: "default"
stopstring | string[]Up to 4 sequences where the API will stop generating further tokens. The returned text will NOT contain the stop sequence.nullable
max_tokensintegerThe maximum number of tokens to generate in the completion. If the token count of your prompt plus max_tokens exceeds the model's context length, the behavior depends on context_length_exceeded_behavior. By default, max_tokens will be lowered to fit in the context window instead of returning an error.nullable
max_completion_tokensintegerAlias for max_tokens. Cannot be specified together with max_tokens.nullable
top_pnumberAn alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered. We generally recommend altering this or temperature but not both. Required range: `0 <= x <= 1` Example: `1`nullable
min_pnumberMinimum probability threshold for token selection. Only tokens with probability >= min_p are considered for selection. This is an alternative to `top_p` and `top_k` sampling. Required range: `0 <= x <= 1`nullable
typical_pnumberTypical-p sampling is an alternative to nucleus sampling. It considers the most typical tokens whose cumulative probability is at most typical_p. Required range: `0 <= x <= 1`nullable
frequency_penaltynumberNumber between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim. Reasonable value is around 0.1 to 1 if the aim is to just reduce repetitive samples somewhat. If the aim is to strongly suppress repetition, then one can increase the coefficients up to 2, but this can noticeably degrade the quality of samples. Negative values can be used to increase the likelihood of repetition. See also `presence_penalty` for penalizing tokens that have at least one appearance at a fixed rate. OpenAI compatible (follows OpenAI's conventions for handling token frequency and repetition penalties). Required range: `-2 <= x <= 2`nullable
presence_penaltynumberNumber between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics. Reasonable value is around 0.1 to 1 if the aim is to just reduce repetitive samples somewhat. If the aim is to strongly suppress repetition, then one can increase the coefficients up to 2, but this can noticeably degrade the quality of samples. Negative values can be used to increase the likelihood of repetition. See also `frequency_penalty` for penalizing tokens at an increasing rate depending on how often they appear. OpenAI compatible (follows OpenAI's conventions for handling token frequency and repetition penalties). Required range: `-2 <= x <= 2`nullable
repetition_penaltynumberApplies a penalty to repeated tokens to discourage or encourage repetition. A value of `1.0` means no penalty, allowing free repetition. Values above `1.0` penalize repetition, reducing the likelihood of repeating tokens. Values between `0.0` and `1.0` reward repetition, increasing the chance of repeated tokens. For a good balance, a value of `1.2` is often recommended. Note that the penalty is applied to both the generated output and the prompt in decoder-only models. Required range: `0 <= x <= 2`nullable
mirostat_targetnumberDefines the target perplexity for the Mirostat algorithm. Perplexity measures the unpredictability of the generated text, with higher values encouraging more diverse and creative outputs, while lower values prioritize predictability and coherence. The algorithm dynamically adjusts the token selection to maintain this target during text generation. If not specified, Mirostat sampling is disabled.nullable
mirostat_lrnumberSpecifies the learning rate for the Mirostat sampling algorithm, which controls how quickly the model adjusts its token distribution to maintain the target perplexity. A smaller value slows down the adjustments, leading to more stable but gradual shifts, while higher values speed up corrections at the cost of potential instability.nullable
seedintegerRandom seed for deterministic sampling.nullable
logprobsinteger | booleanInclude log probabilities in the response. This accepts either a boolean or an integer:nullable
top_logprobsintegerAn integer specifying the number of most likely tokens to return at each token position, each with an associated log probability. Must be between 0 and the deployment's `--max-logprobs` limit (5 by default). When `logprobs` is set, `top_logprobs` can be used to modify how many top log probabilities are returned. If `top_logprobs` is not set, the API will return up to `logprobs` tokens per position.nullable
sampling_maskenumOpt-in sampling mask metadata for generated tokens. When set to `"count"`, each generated token in the new logprobs format includes the number of token logits still eligible for sampling after filters such as top_p and top_k are applied. `"non_zero_list"` additionally returns active token IDs in `sampling_mask`; `"non_zero_buffer"` additionally returns a base64-encoded little-endian uint32 buffer of active token IDs. Non-zero payloads are omitted for positions with more active tokens than 1000.count | non_zero_list | non_zero_buffer · nullable
echobooleanEcho back the prompt in addition to the completion.nullable
echo_lastintegerEcho back the last N tokens of the prompt in addition to the completion. This is useful for obtaining logprobs of the prompt suffix but without transferring too much data. Passing `echo_last=len(prompt)` is the same as `echo=True`nullable
ignore_eosbooleanThis setting controls whether the model should ignore the End of Sequence (EOS) token. When set to `True`, the model will continue generating tokens even after the EOS token is produced. By default, it stops when the EOS token is reached.default: false
context_length_exceeded_behaviorenumWhat to do if the token count of prompt plus `max_tokens` exceeds the model's context window.error | truncate · default: "truncate"
logit_biasobjectModify the likelihood of specified tokens appearing in the completion. Accepts a json object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling.nullable
speculationstring | integer[]Speculative decoding prompt or token IDs to speed up generation.nullable
predictionPredictedOutput | stringOpenAI-compatible predicted output for speculative decoding. Can be a PredictedOutput object or a simple string. Automatically transformed to speculation.nullable
metadataobjectAdditional metadata to store with the request for tracing/distillation.nullable
reasoning_effortenum | integer | booleanControls reasoning behavior for supported models. When enabled, the model's reasoning appears in the `reasoning_content` field of the response, separate from the final answer in `content`.nullable
reasoning_historyenumControls how historical assistant reasoning content is included in the prompt for multi-turn conversations. **Accepted values:** - `null`: Use model/template default behavior (for **GLM-4.7**, the model/template default is `'interleaved'`, i.e. historical reasoning is cleared by default) - `'disabled'`: Strip `reasoning_content` from all messages before prompt construction - `'interleaved'`: Strip `reasoning_content` from messages up to (and including) the last user message - `'preserved'`: Preserve historical `reasoning_content` across the conversation **Model support:** | Model | Default | Supported values | | --- | --- | --- | | Kimi K2.7 | `'preserved'` | `'disabled'`, `'interleaved'`, `'preserved'` | | Kimi K2.6 | `'interleaved'` | `'disabled'`, `'interleaved'`, `'preserved'` | | Kimi K2 Instruct | `'preserved'` | `'disabled'`, `'interleaved'`, `'preserved'` | | MiniMax M2 | `'interleaved'` | `'disabled'`, `'interleaved'` | | GLM-5.2 | `'interleaved'` | `'disabled'`, `'interleaved'`, `'preserved'` | | GLM-4.7 | `'interleaved'` | `'disabled'`, `'interleaved'`, `'preserved'` | | GLM-4.6 | `'interleaved'` | `'disabled'`, `'interleaved'` | | Qwen 3.6 | `'preserved'` | `'disabled'`, `'preserved'` | | DeepSeek V4 | `'interleaved'` | `'interleaved'` | For other models, refer to the model provider's documentation. **Note:** This parameter controls prompt formatting only. To disable reasoning computation entirely, use `reasoning_effort='none'`.disabled | interleaved | preserved · nullable
thinkingThinkingConfigEnabled | ThinkingConfigDi…Configuration for enabling extended thinking (Anthropic-compatible format). This is an alternative to `reasoning_effort` for controlling reasoning behavior.nullable
return_token_idsbooleanReturn token IDs alongside text to avoid retokenization drift.nullable
prompt_token_idsinteger[]Pre-tokenized prompt. When set, the server skips chat-template rendering and tokenization for the input — the supplied token IDs are fed directly to generation. Generation, the response formatter (incl. ``--formatter-backend vllm`` parsers), and tool-call routing all run as usual on the model's output. Intended for upstream gateways (e.g. dynamo) that own the tokenizer and chat template, and want this server only to generate and format. Mutually exclusive with ``messages``: passing both is a 400.nullable
functionsChatCompletionFunction[]Deprecated in OpenAI. Use 'tools' instead. This will be automatically transformed to tools.deprecated
prompt_truncate_lenintegerThe size (in tokens) to which to truncate chat prompts. This includes the system prompt (if any), previous user/assistant messages, and the current user message. Earlier user/assistant messages will be evicted first to fit the prompt into this length. The system prompt is preserved whenever possible and only truncated as a last resort. This should usually be set to a number much smaller << than the model's maximum context size, to allow enough remaining tokens for generating a response. If omitted, you may receive "prompt too long" errors in your responses as conversations grow. Note that even with this set, you may still receive "prompt too long" errors if individual messages (such as a very long system prompt or user message) exceed the model's context window on their own.nullable
parallel_tool_callsbooleanEnable parallel function calling.nullable
safe_tokenizationbooleanWhen true, special tokens in user-provided content are never interpreted as actual special tokens during tokenization. This prevents prompt injection via special token strings (e.g. <|im_start|>, <|User|>). Supported for models using Jinja or HuggingFace chat templates with HuggingFace tokenizers. Explicit true returns an error if the model does not support it, or if combined with custom_chat_template on HuggingFace-backed models. Note: prompt_truncate_len is not applied when safe_tokenization is enabled. When omitted, safe tokenization is enabled by default on a best-effort basis: it is applied only where the model supports it and the request is text-only with no custom_chat_template and no prompt_truncate_len, and otherwise falls back to normal tokenization.nullable
function_callenum | FunctionNameSpecDeprecated in OpenAI. Use 'tool_choice' instead. This will be automatically transformed to tool_choice.nullable · deprecated

Validate a payload

before you spend tokens
shell
$ curl -X POST https://modelschemas.com/v1/validate \
    -d '{"provider":"fireworks","endpointId":"v1/chat/completions","payload":{…}}'
→ {"valid": false, "errors": [{"path": "#", "keyword": "required", …}]}