SambaNova Systems Messages API
The Messages API from SambaNova Systems — 2 operation(s) for messages.
SambaNova Systems Messages API is one of 12 APIs that SambaNova Systems publishes on the APIs.io network, described by a machine-readable OpenAPI specification.
Tagged areas include Message. The published artifact set on APIs.io includes an OpenAPI specification.
This API exposes 2 operations across 2 paths, and defines 76 schemas. It is described by OpenAPI 3.1.0, at version 0.0.1.
Requests are made against a single base URL, https://chat.sambanova.ai/api.
2 operations
2 paths
76 schemas
2 POST
Metadata
The identity and technical contract details declared by the specification.
Authentication & Security 1
SambaNova Systems Messages API declares
1 security scheme
for authenticating requests.
It accepts HTTP bearer tokens (apiKey) (api_key).
By default, every request must be authenticated.
api_key— SambaNova API Key
Paths & Operations 2
Across 2 paths, the API surfaces 2 operations — 2 POST. Each is listed below with its method, path, parameters, and response codes.
Messages 2
POST
/messages
Create a message
POST
/messages/count_tokens
Count tokens for a message request
Schemas 76
The contract defines 76 schemas that model the data the API accepts and returns. The most detailed are MessageCreateRequest (15 properties), Message (10 properties), MessageTool (10 properties), MessageUsage (8 properties). Each schema is shown below with its type and property counts.
MessageThinkingConfig
Controls Anthropic-style extended thinking. In v1: only type:"disabled" is silently accepted as a no-op; type:"enabled" and type:"adaptive" return a 400 invali…
MessageThinkingDisabled
object
Disables Anthropic-style extended thinking. In v1: silently accepted as a no-op
1 property
1 required
MessageCountTokensRequest
object
Request body for POST /messages/counttokens. Returns the input token count for a prompt without generating output. Same prompt shape as MessageCreateRequest mi…
6 properties
2 required
MessageOutputTextEditorCodeExecutionToolResultBlock
object
Anthropic compatibility only — SambaNova does not run server-side text-editor code execution. Never emitted in responses.
3 properties
3 required
MessageInputVideoSourceUrl
object
HTTPS URL pointing to a video.
2 properties
2 required
MessageOutputToolUseBlock
object
Tool call generated by the model.
5 properties
4 required
MessageOutputCodeExecutionToolResultBlock
object
Anthropic compatibility only — SambaNova does not run server-side codeexecution. Never emitted in responses.
3 properties
3 required
MessageInputCodeExecutionToolResultBlock
object
Anthropic compatibility only — SambaNova does not run server-side codeexecution. Echo of a prior Anthropic-served codeexecution tool call; accepted in conversa…
4 properties
3 required
MessageError
object
Inner error object carried inside a MessageErrorResponse. The type value follows Anthropic's published error taxonomy.
2 properties
2 required
MessageInputTextBlock
object
Plain-text segment of a message.
4 properties
2 required
MessageOutputRedactedThinkingBlock
object
Anthropic compatibility only — SambaNova does not produce encrypted thinking output. Never emitted in responses.
2 properties
2 required
MessageThinkingEnabled
object
Enables Anthropic-style extended thinking with a fixed budget. In v1: returns a 400 invalidrequesterror (unsupportedparameter).
2 properties
2 required
MessageContentBlockStopEvent
object
Closes the current content block.
2 properties
2 required
MessageCacheControl
object
Marks the preceding content block (or system text block) as a prompt- cache breakpoint. Marker positions are collected by the adapter; their wiring into the ro…
2 properties
1 required
Message
object
Non-streaming response from POST /messages. Wire-compatible with the official Anthropic Messages API.
10 properties
7 required
MessageOutputWebFetchToolResultBlock
object
Anthropic compatibility only — SambaNova does not run server-side webfetch. Never emitted in responses.
3 properties
3 required
MessageInputServerToolUseBlock
object
Anthropic compatibility only — SambaNova does not run server-side tools. A prior assistant turn that invoked an Anthropic-hosted tool (websearch, codeexecution…
5 properties
4 required
MessageContentBlockTextDelta
object
Incremental text chunk for an open text content block.
2 properties
2 required
MessageInputImageBlock
object
Image content. Only source.type:"base64" is supported in v1; URL sources return 400.
3 properties
2 required
MessageInputRedactedThinkingBlock
object
Anthropic compatibility only — SambaNova does not produce encrypted thinking output. Echo of a prior Anthropic-served response where thinking.display:"omitted"…
2 properties
2 required
MessageOutputTextBlock
object
Plain-text segment of the model's response.
3 properties
2 required
MessageOutputContainerUploadBlock
object
Anthropic compatibility only — SambaNova does not produce containerupload blocks (these come from Anthropic's server-side codeexecution tool). Never emitted in…
2 properties
2 required
MessageContainer
object
Code-execution container reference. Anthropic compatibility only — SambaNova does not run server-side code execution, so this field is never emitted on respons…
2 properties
2 required
MessageToolChoiceNone
object
Forbid the model from calling any tool.
1 property
1 required
MessageStopEvent
object
Final event of the stream. No fields beyond type.
1 property
1 required
MessageUsage
object
Token accounting for the request.
8 properties
2 required
MessageOutputToolSearchToolResultBlock
object
Anthropic compatibility only — SambaNova does not run server-side toolsearch. Never emitted in responses.
3 properties
3 required
MessageInputToolSearchToolResultBlock
object
Anthropic compatibility only — SambaNova does not run server-side toolsearch. Echo of a prior Anthropic-served toolsearch tool call; accepted in conversation h…
4 properties
3 required
MessageInputToolResultContent
Content of a toolresult block. only text is forwarded.
MessageContentBlockInputJsonDelta
object
Incremental fragment of a tooluse block's input JSON. Concatenate successive partialjson strings to reconstruct the full input object.
2 properties
2 required
MessageStartEvent
object
First event of a stream. Carries the initial Message envelope (empty content[], stopreason: null) and token usage from prompt processing.
2 properties
2 required
MessageErrorResponse
object
Top-level error envelope returned by the Messages API on any non-2xx response. Shape matches Anthropic's wire format. This envelope is used ONLY on /messages a…
3 properties
2 required
MessageOutputThinkingBlock
object
Extended-reasoning trace from the model. Emitted by reasoning models.
3 properties
2 required
MessageTool
object
User-defined function tool definition. Only custom function tools are supported (Anthropic's type:"custom" style or the absent-type Beta style). Anthropic-host…
10 properties
1 required
MessageMetadata
object
Free-form metadata attached to the request. Currently only userid Additional fields are accepted but ignored.
1 property
MessageCreateRequest
object
Request body for POST /messages. Wire-compatible with the official Anthropic Messages API. Required fields: model, maxtokens, messages.
15 properties
3 required
MessageInputVideoSourceBase64
object
Inline video data encoded as base64.
3 properties
3 required
MessageInputImageSourceUrl
object
HTTPS URL pointing to an image. Returns 400 in v1 — URL fetching is blocked. Use type:"base64" instead.
2 properties
2 required
MessageOutputContentBlock
Typed content block in the model's response.
MessageInputDocumentBlock
object
PDF or document content. Returns 400 — no document-extraction pipeline available.
6 properties
2 required
MessageStreamEvent
Discriminated union of SSE events emitted during a streaming POST /messages call. Event order: messagestart - (contentblockstart - contentblockdelta×N - conten…
MessageContentBlockThinkingDelta
object
Incremental thinking chunk for an open thinking block. Emitted by reasoning models.
2 properties
2 required
MessagePingEvent
object
Keepalive heartbeat. May appear at any point in the stream.
1 property
1 required
MessageSystemPrompt
System prompt for the conversation. Accepts either a single string (most common) or an array of text blocks (used when individual segments need cachecontrol ma…
MessageDeltaUsage
object
Final token accounting emitted in the closing messagedelta event of a stream.
5 properties
1 required
MessageOutputServerToolUseBlock
object
Anthropic compatibility only — SambaNova does not run server-side tools. Never emitted in responses; defined for Anthropic SDK type-parity.
4 properties
4 required
MessageDeltaEvent
object
Penultimate event of the stream. Carries the final stopreason, optional stopsequence, and final usage counts.
3 properties
3 required
MessageOutputWebSearchToolResultBlock
object
Anthropic compatibility only — SambaNova does not run server-side websearch. Never emitted in responses.
3 properties
3 required
MessageStopDetails
object
Refusal stop details. Anthropic compatibility only — refusal is never emitted as a stopreason by SambaNova (content filtering is not exposed at the API layer).
2 properties
1 required
MessageInputMessage
object
A turn in the conversation.
2 properties
2 required
MessageInputSearchResultBlock
object
Inline search result content. In v1 the title, source, and content[] text are extracted into a text block; citations are dropped.
6 properties
1 required
MessageInputBashCodeExecutionToolResultBlock
object
Anthropic compatibility only — SambaNova does not run server-side bash code execution. Echo of a prior Anthropic-served bash tool call; accepted in conversatio…
4 properties
3 required
MessageInputImageSourceBase64
object
Inline image data encoded as base64.
3 properties
3 required
MessageInputThinkingBlock
object
Extended-reasoning trace from a prior assistant turn.
3 properties
3 required
MessageInputWebFetchToolResultBlock
object
Anthropic compatibility only — SambaNova does not run server-side webfetch. Echo of a prior Anthropic-served webfetch tool call; accepted in conversation histo…
4 properties
3 required
MessageToolChoiceAny
object
Require the model to call one of the provided tools.
2 properties
1 required
MessageInputVideoBlock
object
Video content.
3 properties
2 required
MessageInputWebSearchToolResultBlock
object
Anthropic compatibility only — SambaNova does not run server-side websearch. Echo of a prior Anthropic-served websearch tool call; accepted in conversation his…
4 properties
3 required
MessageToolChoice
How the model should choose from the provided tools.
MessageStreamErrorEvent
object
Streamed error envelope. Terminates the stream.
2 properties
2 required
MessageContentBlockSignatureDelta
object
Signature for an open thinking block. Emitted at the end of a thinking stream (paired with the closing contentblockstop); the signature value may be an empty s…
2 properties
2 required
MessageSystemTextBlock
object
A text segment within a structured system prompt array. Multiple text blocks are concatenated (with newlines) and prepended to the conversation as a role: syst…
4 properties
2 required
MessageInputContainerUploadBlock
object
Anthropic compatibility only — SambaNova does not produce containerupload blocks (these come from Anthropic's server-side codeexecution tool). Accepted in conv…
3 properties
2 required
MessageStopReason
string
Reason the model stopped generating. SambaNova emits endturn, maxtokens, tooluse, and stopsequence. The remaining values are defined for Anthropic SDK type-par…
MessageToolChoiceAuto
object
Let the model decide whether and which tool to use.
2 properties
1 required
MessageThinkingAdaptive
object
Enables Anthropic-style adaptive extended thinking. In v1: returns a 400 invalidrequesterror (unsupportedparameter).
2 properties
1 required
MessageContentBlockDelta
Incremental update to an open content block.
MessageOutputBashCodeExecutionToolResultBlock
object
Anthropic compatibility only — SambaNova does not run server-side bash code execution. Never emitted in responses.
3 properties
3 required
MessageToolChoiceTool
object
Force the model to call a specific tool by name.
3 properties
2 required
MessageInputContentBlock
Typed content block inside a message's content array.
MessageInputToolUseBlock
object
A prior assistant turn that invoked a tool.
5 properties
4 required
MessageContentBlockDeltaEvent
object
Incremental update to the currently open content block.
3 properties
3 required
MessageContentBlockStartEvent
object
Opens a new content block. One per block in content[].
3 properties
3 required
MessageCountTokensResponse
object
Token count for the supplied prompt.
1 property
1 required
MessageInputTextEditorCodeExecutionToolResultBlock
object
Anthropic compatibility only — SambaNova does not run server-side text-editor code execution. Echo of a prior Anthropic-served text-editor tool call; accepted…
4 properties
3 required
MessageInputToolResultBlock
object
Result of a prior tool call.
5 properties
2 required
Specification
The full machine-readable OpenAPI contract behind this narrative.
Source
More from SambaNova Systems 11
Other APIs SambaNova Systems publishes across the network.
SambaNova Systems Audio API
SambaNova Systems Chat completions API
SambaNova Systems Completions API
SambaNova Systems Embeddings API
SambaNova Systems File Management API
SambaNova Systems Main Agent API
SambaNova Systems Main Agent Interactive API
SambaNova Systems Models API
SambaNova Systems Responses API
SambaNova Systems Subagents API
SambaNova Systems Subagents Interactive API
#Message
Where this information came from
This is an independent, third-party profile of SambaNova Systems Messages API, published by API Evangelist. We do not operate, host, resell, or support these APIs, and we are not affiliated with or endorsed by the company unless stated above. Everything here is built from publicly available information — the company's own site, developer portal, documentation, public repositories, and the specifications it publishes for public use. Nothing is obtained by breaching a system, defeating an access control, or using credentials.
The Kin Score and Agent Readiness rating are independently calculated assessments of a company's public API artifacts, scored against a published rubric. They are not certifications, endorsements, security assessments, or audits.
Corrections, re-scores, and removal are free — no partnership or purchase required, and you do not need to justify the request. A removed company is recorded as unrated, never scored zero for having asked. Acknowledgement within one business day; removal within two.
info@apievangelist.com
·
Read the full data-sourcing policy →
On a security or compliance team? Put security in the subject line and
you will get a person, not a form — we will tell you exactly which public URLs this profile was built from.