Together AI MCP by usefulapi is an MCP server with 23 tools an agent can call, among them together_cancel_batch, together_cancel_fine_tune, together_chat_completion. It answers at https://together-ai.usefulapi.io/mcp over streamable HTTP, and it listed its tools to an anonymous client, so no sign-in was needed for that. Our probes reached it 4 of 4 times since 29 September 2026; the last check, on 15 November 2026, succeeded.
Run Together AI chat, embeddings and images; manage fine-tunes, batches and endpoints.
Description as the publisher wrote it.
| Tool | What it does |
|---|---|
together_cancel_batch | Cancel a batch job that has not finished. Together: POST /batches/{id}/cancel. Inputs: batch_id |
together_cancel_fine_tune | Cancel a running fine-tuning job. Cannot be resumed, but a new job can continue from its last checkpoint via from_checkpoint. Together: POST /fine-tunes/{id}/cancel. Inputs: fine_tune_id |
together_chat_completion | Run a chat completion on a Together model (billed per token). Non-streaming. For a dedicated endpoint pass its `<project_slug>/<endpoint_slug>` as the model. Together: POST /chat/completions. Inputs: model, messages, max_tokens, temperature, top_p, top_k |
together_create_batch | Start an asynchronous batch job over an uploaded JSONL input file (purpose batch-api), at a discount to real-time inference. Together: POST /batches. Inputs: input_file_id, endpoint, model_id, completion_window, priority |
together_create_embeddings | Generate vector embeddings for one or more texts (billed per token). Together: POST /embeddings. Inputs: model, input |
together_create_endpoint | Deploy a model on dedicated GPUs. The endpoint STARTS AUTOMATICALLY and bills per minute of uptime until stopped — set inactive_timeout to auto-stop it, and use together_list_hardware for valid hardware ids. Together: POST /endpoints. Inputs: model, hardware, autoscaling, display_name, availability_zone, inactive_timeout |
together_create_fine_tune | Start a fine-tuning job on an uploaded training file (billed per token processed). Stop it with together_cancel_fine_tune. Together: POST /fine-tunes. Inputs: model, training_file, validation_file, suffix, n_epochs, n_checkpoints |
together_generate_image | Generate images from a prompt (billed per image/megapixel). Returns image URLs by default rather than base64, to keep responses small. Together: POST /images/generations. Inputs: model, prompt, negative_prompt, width, height, n |
together_get_batch | Fetch one batch job: status (VALIDATING, IN_PROGRESS, COMPLETED, FAILED, EXPIRED, CANCELLED), progress, and the output_file_id / error_file_id once done. Together: GET /batches/{id}. Inputs: batch_id |
together_get_endpoint | Fetch one dedicated endpoint: state, model, hardware, autoscaling bounds and display name. Together: GET /endpoints/{endpointId}. Inputs: endpoint_id |
together_get_file | Fetch one file's metadata, including its processing_status and validation_report (why a fine-tune training file was rejected). Together: GET /files/{id}. Inputs: file_id |
together_get_fine_tune | Fetch one fine-tuning job: status, progress, hyperparameters, token counts, cost and the output model name. Together: GET /fine-tunes/{id}. Inputs: fine_tune_id |
together_list_batches | List batch inference jobs with status, progress, model and input/output/error file ids. Together: GET /batches. |
together_list_endpoints | List endpoints with model, owner and state (PENDING, STARTING, STARTED, STOPPING, STOPPED, ERROR). Use mine=true and type=dedicated to see what is running on your account and billing by the minute. Together: GET /endpoints. Inputs: type, usage_type, mine |
together_list_evaluations | List LLM-as-a-judge evaluation jobs (classify, score, compare) with status, parameters and results once completed. Together: GET /evaluation. Inputs: status, limit |
together_list_files | List uploaded data files (fine-tune, eval and batch-api inputs, plus job outputs) with size, type, purpose and validation status. Together: GET /files. |
together_list_fine_tune_events | List the event log of one fine-tuning job (queued, started, checkpoint saved, epoch completed, errors). The first place to look when a job failed. Together: GET /fine-tunes/{id}/events. Inputs: fine_tune_id |
together_list_fine_tunes | List fine-tuning jobs with status, base model, output model name and training settings. Together: GET /fine-tunes. |
together_list_hardware | List hardware configurations for dedicated endpoints with GPU type/count/memory and price in cents per minute. Pass a model to get only compatible configurations with live availability. Together: GET /hardware. Inputs: model |
together_list_models | List Together's models with type (chat, language, code, image, embedding, moderation, rerank), context length, organization, license and per-token pricing. Together: GET /models. Inputs: dedicated |
together_start_endpoint | Start a stopped dedicated endpoint. It bills per minute of uptime until stopped. Reversible with together_stop_endpoint. Together: PATCH /endpoints/{endpointId} with state=STARTED. Inputs: endpoint_id |
together_stop_endpoint | Stop a running dedicated endpoint, which stops its per-minute billing. Requests to it fail until it is started again. Together: PATCH /endpoints/{endpointId} with state=STOPPED. Inputs: endpoint_id |
together_whoami | Identify the API key: its organization, project and project slug. The project slug forms the `<project_slug>/<endpoint_slug>` model name for dedicated-endpoint inference. A cheap way to confirm the key works. Together: GET /whoami. |
Every block below points at https://together-ai.usefulapi.io/mcp. Pick your client and copy.
claude mcp add --transport http together-ai-mcp-by-usefulapi https://together-ai.usefulapi.io/mcp
{
"mcpServers": {
"together-ai-mcp-by-usefulapi": {
"command": "npx",
"args": ["-y", "mcp-remote", "https://together-ai.usefulapi.io/mcp"]
}
}
}{
"mcpServers": {
"together-ai-mcp-by-usefulapi": {
"url": "https://together-ai.usefulapi.io/mcp"
}
}
}{
"servers": {
"together-ai-mcp-by-usefulapi": {
"type": "http",
"url": "https://together-ai.usefulapi.io/mcp"
}
}
}{
"context_servers": {
"together-ai-mcp-by-usefulapi": {
"url": "https://together-ai.usefulapi.io/mcp"
}
}
}{
"mcpServers": {
"together-ai-mcp-by-usefulapi": {
"serverUrl": "https://together-ai.usefulapi.io/mcp"
}
}
}{
"mcpServers": {
"together-ai-mcp-by-usefulapi": {
"url": "https://together-ai.usefulapi.io/mcp"
}
}
}name: Together AI MCP by usefulapi
version: 0.0.1
schema: v1
mcpServers:
- name: Together AI MCP by usefulapi
type: streamable-http
url: https://together-ai.usefulapi.io/mcp{
"mcpServers": {
"together-ai-mcp-by-usefulapi": {
"httpUrl": "https://together-ai.usefulapi.io/mcp"
}
}
}{
"mcpServers": {
"together-ai-mcp-by-usefulapi": {
"type": "http",
"url": "https://together-ai.usefulapi.io/mcp"
}
}
}{
"mcpServers": {
"together-ai-mcp-by-usefulapi": {
"url": "https://together-ai.usefulapi.io/mcp"
}
}
}goose://extension?url=https%3A%2F%2Ftogether-ai.usefulapi.io%2Fmcp&type=streamable_http&id=together-ai-mcp-by-usefulapi&name=Together%20AI%20MCP%20by%20usefulapi
It answered our last check on 15 November 2026 in 102 ms. Over 4 checks since 29 September 2026 it answered 4. Status here comes from Neuronto's own probes, not from the publisher.
Yes. It is published by usefulapi.io and served from together-ai.usefulapi.io, the vendor's own domain: https://together-ai.usefulapi.io/mcp.
23, as returned by the server's own tools/list on 18 November 2026: together_cancel_batch, together_cancel_fine_tune, together_chat_completion, together_create_batch, together_create_embeddings, together_create_endpoint, together_create_fine_tune, together_generate_image, and 15 more.
It answered our last check on 15 November 2026 in 102 ms. Over 4 checks since 29 September 2026 it answered 4.
Put the server under mcpServers in ~/.cursor/mcp.json with the URL https://together-ai.usefulapi.io/mcp. The exact JSON is above, with the same setup for Claude Code, Claude Desktop, VS Code, Windsurf, Zed, Gemini CLI and the rest.
It listed its tools to an anonymous client, so reading the tool list needs no key. Individual tools can still ask for credentials in their own inputs.
Publisher: io.usefulapi. Endpoint host: together-ai.usefulapi.io. ARD identifier: urn:air:io.usefulapi:mcp:together-ai.
First indexed 29 September 2026. The tools are exactly what the server returned; nothing on this page
is inferred from its name. Ranking in Neuronto search is never for sale.