Connectors · AI
Together AI MCP server for AI agents.
actionlake is an MCP gateway: one remote MCP server that connects Claude and other MCP clients to Together AI and the rest of your systems. It gives your agent 183 Together AI actions (101 reads, 57 writes and 25 that wait for a confirmation) and holds the Together AI credential itself, so the agent never does. It paces Together AI's limits and records every call.
100,000 free calls a month. No card, no email, no password: press Start a free sandbox on the next screen and keep the key it shows you.
claude mcp add --transport http actionlake https://connect.actionlake.com/mcpdo { action: "together_ai.deployments.list", intent: "check which Together AI account is connected"}# Get the list of deployments · read · 1 upstream call- 183actionsin one signed file
- 101readlook, never change
- 57writecreate and update
- 25wait for a yesdestructive, confirmed first
Set up
Connect Together AI in three steps.
- 01
Add actionlake to Claude.
Run the command once in Claude Code; its consent page can create a free sandbox: 100,000 calls a month, no card. In the Claude app, add a custom connector with the URL below and paste your key on the consent page. Creating a new sandbox from the Claude app is not yet tested by us.
https://connect.actionlake.com/mcp - 02
Connect Together AI.
API key. No consent screen: your agent can connect it for you, but the key then passes through your chat. To keep it out of the model's context, paste it in the console instead. actionlake stores what you paste write-only, sealed under your tenant's key.
api_keyTogether AI API key (api.together.ai > Settings > API Keys)
account { op: "connections.requirements", service: "together_ai" } # free: exactly what to paste, and what is found for you - 03
Make the first call. Check before a write.
Ask in plain words; Claude picks the action. Before anything changes,
checkvalidates the call and prices it, for free.do { action: "together_ai.deployments.list" }# get the list of deployments · read · 1 upstream callcheck { action: "together_ai.deployments.create", input: { … } }# free: inputs, permissions, cost and confirmations, nothing sent
Together AI at a glance
- Sign-in
- API key. No consent screen: your agent can connect it for you, but the key then passes through your chat. To keep it out of the model's context, paste it in the console instead.
- You provide
api_key- Limits we pace to
- 60 calls a minute, per connection
- API covered
- Together AI API 2.0.0
api.together.ai - Built from
- Together AI's documentation, read 23 October 2026docs.together.aidocs.together.ai
- Service file
- Version 1.0.0, signed (Ed25519)
- Cached reads
- 87 of 101 reads, for up to 5 minutes; a write clears what it affects
- Shared entities
- 2, so ask can line Together AI up with your other systems
- Data
- Processed and stored in Sydney, Australia. No US region yet
Capabilities
What an AI agent can do in Together AI.
183 actions in 74 areas. Every one carries its effect, so a key can allow the reads and hold the rest.
- 101readlook, never change
- 57writecreate and update
- 25destructivewait for a confirmation
Projects endpoints rollouts
103 read · 5 write · 2 destructive
Create a rollout · List endpoint rollouts · Get a rollout · Delete a rollout
Fine tunes
74 read · 1 write · 2 destructive
Create job · List all jobs · List job · Delete a fine-tune job
RL training sessions operations
70 read · 7 write · 0 destructive
Custom forward-backward pass · Forward-backward pass · Create inference checkpoint · Optimizer step
Compute clusters instances remediations
62 read · 3 write · 1 destructive
compute_clusters_instances_remediations.create · List remediations · compute_clusters_instances_remediations.get · compute_clusters_instances_remediations.approve
Deployments
62 read · 3 write · 1 destructive
Get the list of deployments · Create a new deployment · Get a deployment by ID or name · Update a deployment
Compute clusters
52 read · 2 write · 1 destructive
Create a GPU cluster · List all GPU clusters · Get GPU cluster by cluster ID · Update a GPU cluster
Compute clusters addons
52 read · 2 write · 1 destructive
compute_clusters_addons.create · compute_clusters_addons.list · compute_clusters_addons.get · compute_clusters_addons.update
Compute clusters storage volumes
52 read · 2 write · 1 destructive
Create a shared volume · List all shared volumes · Get a shared volume by ID · Update a shared volume
66 more areas · 132 actions
- Deployments secrets 5
- Deployments storage volumes 5
- Projects endpoints 5
- Projects endpoints ab experiments 5
- Projects endpoints deployments 5
- Projects endpoints deployments adapters 5
- Projects endpoints shadow experiments 5
- Projects endpoints shadow experiments targets 5
- Projects models 5
- RL model resources 5
- RL training sessions 5
- Batches 4
- Evaluation 3
- Files 3
- Projects models uploads 3
- Queue jobs 3
- Shaping 3
- Deployments revisions 2
- Projects configs 2
- Projects models revisions 2
- Projects placement profiles 2
- Public inference instance types 2
- RL checkpoints 2
- Shaping FP4 jobs 2
- Supported models 2
- Videos 2
- Billing usage 1
- Chat completions 1
- Completions 1
- Compute passive health check alerts 1
- Compute regions 1
- Deployments logs 1
- Deployments storage files 1
- Embeddings 1
- Evaluation models 1
- Evaluation status 1
- Fine tunes checkpoints 1
- Fine tunes download tokenized dataset 1
- Fine tunes events 1
- Fine tunes metrics 1
- Fine tunes models limits 1
- Fine tunes models supported 1
- Image generations 1
- Organizations endpoints 1
- Organizations models 1
- Projects 1
- Projects endpoints analytics 1
- Projects endpoints events 1
- Projects models files 1
- Projects models uploads events 1
- Public endpoints 1
- Queue 1
- Queue metrics 1
- Rerank 1
- RL checkpoints download urls 1
- RL supported models 1
- RL training sessions operations custom forward backward 1
- RL training sessions operations forward backward 1
- RL training sessions operations inference checkpoint 1
- RL training sessions operations optim step 1
- RL training sessions operations sample 1
- RL training sessions operations training checkpoint 1
- RL training sessions operations weights sync 1
- Shaping events 1
- Voices 1
- Whoami 1
Top actions
The Together AI API, shaped for agents.
The ids your agent types, what each does, and the one name it shares with other systems. 32 of Together AI's 32 action names are shared with at least one other connector, so an agent learns them once.
| Action | What it does | Effect | Same name in |
|---|---|---|---|
together_ai.deployments.listGet the list of deployments | Get the list of deployments | read | infra.deployment.list8 connectors |
together_ai.files.listList all files | List all files | read | content.file.list77 connectors |
together_ai.deployments.createCreate a new deployment | Create a new deployment | write | infra.deployment.create7 connectors |
together_ai.files.getRetrieve file metadata | Retrieve file metadata | read | content.file.get74 connectors |
together_ai.deployments.deleteDelete a deployment | Delete a deployment | destructive | infra.deployment.delete7 connectors |
together_ai.compute_regions.listList regions and corresponding supported driver versions | List regions and corresponding supported driver versions | read | infra.region.list16 connectors |
together_ai.deployments_storage_volumes.listGet the list of project volumes | Get the list of project volumes | read | infra.volume.list10 connectors |
together_ai.batches.listList batch jobs | List batch jobs | read | automation.background_job.list34 connectors |
together_ai.deployments_storage_volumes.createCreate a new volume | Create a new volume | write | infra.volume.create8 connectors |
together_ai.whoami.getIdentify the caller | Identify the caller | read | identity.current_user.get307 connectors |
together_ai.batches.createCreate a batch job | Create a batch job | write | automation.background_job.create29 connectors |
together_ai.compute_clusters.createCreate a GPU cluster | Create a GPU cluster | write | infra.container_cluster.create6 connectors |
Try it
Prompts to paste into Claude.
Each one names the Together AI actions it will call, so you know what it reads and what it changes before you send it.
- Readread
Which Together AI files changed this week, and who changed them?
calls
together_ai.files.list - Writewrite
Create a new deployment in Together AI. Show me exactly what you will send before you save it.
calls
together_ai.deployments.create
Governance for Together AI
Destructive actions wait for a yes.
25 Together AI actions are destructive. These are all of them:
- Delete a deployment
together_ai.deployments.delete - Delete a file
together_ai.files.delete - Delete a volume
together_ai.deployments_storage_volumes.delete - Delete a model
together_ai.projects_models.delete - Delete GPU cluster by cluster ID
together_ai.compute_clusters.delete - Cancel a batch job
together_ai.batches.cancel - Delete a fine-tune job
together_ai.fine_tunes.delete - Delete an endpoint
together_ai.projects_endpoints.delete - Delete a secret
together_ai.deployments_secrets.delete - compute_clusters_addons.delete
together_ai.compute_clusters_addons.delete - Delete a rollout
together_ai.projects_endpoints_rollouts.delete - Delete a deployment
together_ai.projects_endpoints_deployments.delete - Delete a shared volume by ID
together_ai.compute_clusters_storage_volumes.delete - Delete an A/B experiment
together_ai.projects_endpoints_ab_experiments.delete - Delete a shadow experiment
together_ai.projects_endpoints_shadow_experiments.delete - Remove a deployment adapter
together_ai.projects_endpoints_deployments_adapters.delete - Delete a shadow experiment target
together_ai.projects_endpoints_shadow_experiments_targets.delete - Clear a model's pending jobs
together_ai.queue.clear - Cancel shaping job
together_ai.shaping.cancel - Cancel job
together_ai.fine_tunes.cancel - Cancel a queued job
together_ai.queue_jobs.cancel - Stop model resources
together_ai.rl_model_resources.stop - Stop training session
together_ai.rl_training_sessions.stop - Cancel a rollout
together_ai.projects_endpoints_rollouts.cancel - compute_clusters_instances_remediations.cancel
together_ai.compute_clusters_instances_remediations.cancel
A destructive action runs only with confirm: true. Under the human policy the agent cannot confirm for itself: the run stops until a named person approves, in the console or from a signed link in an email. The cap on a key is refused before the call is sent, never after.
Delete a deployment?
intent: the user asked me to delete a deployment
account { op: "keys.create", name: "together-ai-agent", allowed_actions: ["together_ai.*"], cap_calls: 2000, confirm_policy: "human_for_writes", approval_rules: [ { match: "together_ai.deployments.delete", approver: "[email protected]" }, { match: "together_ai.files.delete", approver: "[email protected]" }, { match: "together_ai.deployments_storage_volumes.delete", approver: "[email protected]" } ]}Cost
What Together AI costs on actionlake.
Every Together AI action is one upstream call, the only thing actionlake bills. Searching, check, cost, account and any call Together AI refused with 401, 403, 429 or a 5xx are free. Paid plans start at US$49 a month for 250,000 calls, with overage from US$0.30 down to US$0.18 per 1,000.
- 1upstream call per actionall 183 Together AI actions
- 49actions pageeach page of a long list is one call
- 87reads can be cacheda cache hit costs nothing
- 100,000calls a month, freepaid plans from US$49 a month for 250,000
Action index
Every Together AI action.
183 actions by area: the id your agent calls, what it does, and its effect.
Projects endpoints rollouts10
together_ai.projects_endpoints_rollouts.cancelCancel a rolloutdestructivetogether_ai.projects_endpoints_rollouts.createCreate a rolloutwritetogether_ai.projects_endpoints_rollouts.deleteDelete a rolloutdestructivetogether_ai.projects_endpoints_rollouts.getGet a rolloutreadtogether_ai.projects_endpoints_rollouts.listList endpoint rolloutsreadtogether_ai.projects_endpoints_rollouts.pausePause a rolloutwritetogether_ai.projects_endpoints_rollouts.preview_defaultsPreview rollout defaultsreadtogether_ai.projects_endpoints_rollouts.promotePromote a rolloutwritetogether_ai.projects_endpoints_rollouts.resumeResume a rolloutwritetogether_ai.projects_endpoints_rollouts.startStart a rolloutwrite
Fine tunes7
together_ai.fine_tunes.cancelCancel jobdestructivetogether_ai.fine_tunes.createCreate jobwritetogether_ai.fine_tunes.deleteDelete a fine-tune jobdestructivetogether_ai.fine_tunes.estimate_priceEstimate pricereadtogether_ai.fine_tunes.getList jobreadtogether_ai.fine_tunes.listList all jobsreadtogether_ai.fine_tunes.previewPreview tokenized dataread
RL training sessions operations7
together_ai.rl_training_sessions_operations.custom_forward_backwardCustom forward-backward passwritetogether_ai.rl_training_sessions_operations.forward_backwardForward-backward passwritetogether_ai.rl_training_sessions_operations.inference_checkpointCreate inference checkpointwritetogether_ai.rl_training_sessions_operations.optim_stepOptimizer stepwritetogether_ai.rl_training_sessions_operations.sampleSamplewritetogether_ai.rl_training_sessions_operations.training_checkpointSave training checkpointwritetogether_ai.rl_training_sessions_operations.weights_syncWeights syncwrite
Compute clusters instances remediations6
together_ai.compute_clusters_instances_remediations.approvecompute_clusters_instances_remediations.approvewritetogether_ai.compute_clusters_instances_remediations.cancelcompute_clusters_instances_remediations.canceldestructivetogether_ai.compute_clusters_instances_remediations.createcompute_clusters_instances_remediations.createwritetogether_ai.compute_clusters_instances_remediations.getcompute_clusters_instances_remediations.getreadtogether_ai.compute_clusters_instances_remediations.listList remediationsreadtogether_ai.compute_clusters_instances_remediations.rejectcompute_clusters_instances_remediations.rejectwrite
Deployments6
together_ai.deployments.createCreate a new deploymentwritetogether_ai.deployments.deleteDelete a deploymentdestructivetogether_ai.deployments.getGet a deployment by ID or namereadtogether_ai.deployments.listGet the list of deploymentsreadtogether_ai.deployments.rollbackRoll back a deployment to a previous revisionwritetogether_ai.deployments.updateUpdate a deploymentwrite
Compute clusters5
together_ai.compute_clusters.createCreate a GPU clusterwritetogether_ai.compute_clusters.deleteDelete GPU cluster by cluster IDdestructivetogether_ai.compute_clusters.getGet GPU cluster by cluster IDreadtogether_ai.compute_clusters.listList all GPU clustersreadtogether_ai.compute_clusters.updateUpdate a GPU clusterwrite
Compute clusters addons5
together_ai.compute_clusters_addons.createcompute_clusters_addons.createwritetogether_ai.compute_clusters_addons.deletecompute_clusters_addons.deletedestructivetogether_ai.compute_clusters_addons.getcompute_clusters_addons.getreadtogether_ai.compute_clusters_addons.listcompute_clusters_addons.listreadtogether_ai.compute_clusters_addons.updatecompute_clusters_addons.updatewrite
Compute clusters storage volumes5
together_ai.compute_clusters_storage_volumes.createCreate a shared volumewritetogether_ai.compute_clusters_storage_volumes.deleteDelete a shared volume by IDdestructivetogether_ai.compute_clusters_storage_volumes.getGet a shared volume by IDreadtogether_ai.compute_clusters_storage_volumes.listList all shared volumesreadtogether_ai.compute_clusters_storage_volumes.updateUpdate a shared volumewrite
Deployments secrets5
together_ai.deployments_secrets.createCreate a new secretwritetogether_ai.deployments_secrets.deleteDelete a secretdestructivetogether_ai.deployments_secrets.getGet a secret by ID or namereadtogether_ai.deployments_secrets.listGet the list of project secretsreadtogether_ai.deployments_secrets.updateUpdate a secretwrite
Deployments storage volumes5
together_ai.deployments_storage_volumes.createCreate a new volumewritetogether_ai.deployments_storage_volumes.deleteDelete a volumedestructivetogether_ai.deployments_storage_volumes.getGet a volume by ID or namereadtogether_ai.deployments_storage_volumes.listGet the list of project volumesreadtogether_ai.deployments_storage_volumes.updateUpdate a volumewrite
Projects endpoints5
together_ai.projects_endpoints.createCreate an endpointwritetogether_ai.projects_endpoints.deleteDelete an endpointdestructivetogether_ai.projects_endpoints.getGet an endpointreadtogether_ai.projects_endpoints.listList endpointsreadtogether_ai.projects_endpoints.updateUpdate an endpointwrite
Projects endpoints ab experiments5
together_ai.projects_endpoints_ab_experiments.createCreate an A/B experimentwritetogether_ai.projects_endpoints_ab_experiments.deleteDelete an A/B experimentdestructivetogether_ai.projects_endpoints_ab_experiments.getGet an A/B experimentreadtogether_ai.projects_endpoints_ab_experiments.listList A/B experimentsreadtogether_ai.projects_endpoints_ab_experiments.updateUpdate an A/B experimentwrite
Projects endpoints deployments5
together_ai.projects_endpoints_deployments.createCreate a deploymentwritetogether_ai.projects_endpoints_deployments.deleteDelete a deploymentdestructivetogether_ai.projects_endpoints_deployments.getGet a deploymentreadtogether_ai.projects_endpoints_deployments.listList deploymentsreadtogether_ai.projects_endpoints_deployments.updateUpdate a deploymentwrite
Projects endpoints deployments adapters5
together_ai.projects_endpoints_deployments_adapters.createAdd a deployment adapterwritetogether_ai.projects_endpoints_deployments_adapters.deleteRemove a deployment adapterdestructivetogether_ai.projects_endpoints_deployments_adapters.getGet a deployment adapterreadtogether_ai.projects_endpoints_deployments_adapters.listList deployment adaptersreadtogether_ai.projects_endpoints_deployments_adapters.updateUpdate a deployment adapterwrite
Projects endpoints shadow experiments5
together_ai.projects_endpoints_shadow_experiments.createCreate a shadow experimentwritetogether_ai.projects_endpoints_shadow_experiments.deleteDelete a shadow experimentdestructivetogether_ai.projects_endpoints_shadow_experiments.getGet a shadow experimentreadtogether_ai.projects_endpoints_shadow_experiments.listList shadow experimentsreadtogether_ai.projects_endpoints_shadow_experiments.updateUpdate a shadow experimentwrite
Projects endpoints shadow experiments targets5
together_ai.projects_endpoints_shadow_experiments_targets.createCreate a shadow experiment targetwritetogether_ai.projects_endpoints_shadow_experiments_targets.deleteDelete a shadow experiment targetdestructivetogether_ai.projects_endpoints_shadow_experiments_targets.getGet a shadow experiment targetreadtogether_ai.projects_endpoints_shadow_experiments_targets.listList shadow experiment targetsreadtogether_ai.projects_endpoints_shadow_experiments_targets.updateUpdate a shadow experiment targetwrite
Projects models5
together_ai.projects_models.createCreate a modelwritetogether_ai.projects_models.deleteDelete a modeldestructivetogether_ai.projects_models.getGet a modelreadtogether_ai.projects_models.listList project modelsreadtogether_ai.projects_models.updateUpdate a modelwrite
RL model resources5
together_ai.rl_model_resources.createCreate model resourceswritetogether_ai.rl_model_resources.estimate_costEstimate on-demand model resource costreadtogether_ai.rl_model_resources.getGet model resourcesreadtogether_ai.rl_model_resources.listList model resourcesreadtogether_ai.rl_model_resources.stopStop model resourcesdestructive
RL training sessions5
together_ai.rl_training_sessions.createCreate training sessionwritetogether_ai.rl_training_sessions.getGet training sessionreadtogether_ai.rl_training_sessions.listList training sessionsreadtogether_ai.rl_training_sessions.stopStop training sessiondestructivetogether_ai.rl_training_sessions.updateUpdate a training sessionwrite
Batches4
together_ai.batches.cancelCancel a batch jobdestructivetogether_ai.batches.createCreate a batch jobwritetogether_ai.batches.getGet a batch jobreadtogether_ai.batches.listList batch jobsread
Evaluation3
together_ai.evaluation.createCreate an evaluation jobwritetogether_ai.evaluation.getGet evaluation job detailsreadtogether_ai.evaluation.listGet all evaluation jobsread
Files3
together_ai.files.deleteDelete a filedestructivetogether_ai.files.getRetrieve file metadatareadtogether_ai.files.listList all filesread
Projects models uploads3
together_ai.projects_models_uploads.createCreate a remote model uploadwritetogether_ai.projects_models_uploads.getGet a remote model uploadreadtogether_ai.projects_models_uploads.listList remote model uploadsread
Queue jobs3
together_ai.queue_jobs.cancelCancel a queued jobdestructivetogether_ai.queue_jobs.getGet job statusreadtogether_ai.queue_jobs.submitSubmit a queued jobwrite
Shaping3
together_ai.shaping.cancelCancel shaping jobdestructivetogether_ai.shaping.getGet shaping jobreadtogether_ai.shaping.listList shaping jobsread
Deployments revisions2
together_ai.deployments_revisions.getGet the full configuration of a specific revisionreadtogether_ai.deployments_revisions.listList revision history events of a deploymentread
Projects configs2
together_ai.projects_configs.getGet a model configurationreadtogether_ai.projects_configs.listList model configurationsread
Projects models revisions2
together_ai.projects_models_revisions.getGet a model volume revisionreadtogether_ai.projects_models_revisions.listList model revisionsread
Projects placement profiles2
together_ai.projects_placement_profiles.getGet a placement profilereadtogether_ai.projects_placement_profiles.listList placement profilesread
Public inference instance types2
together_ai.public_inference_instance_types.getGet an inference instance typereadtogether_ai.public_inference_instance_types.listList inference instance typesread
RL checkpoints2
together_ai.rl_checkpoints.getGet checkpointreadtogether_ai.rl_checkpoints.listList checkpointsread
Shaping FP4 jobs2
together_ai.shaping_fp4_jobs.createPrepare a fine-tuned model for FP4 inferencewritetogether_ai.shaping_fp4_jobs.estimateEstimate FP4 preparation costread
Supported models2
together_ai.supported_models.getGet a supported modelreadtogether_ai.supported_models.listList supported modelsread
Videos2
together_ai.videos.createCreate videowritetogether_ai.videos.getFetch video metadataread
Billing usage1
together_ai.billing_usage.listGet billing usageread
Chat completions1
together_ai.chat_completions.createCreate chat completionwrite
Completions1
together_ai.completions.createCreate completionwrite
Compute passive health check alerts1
together_ai.compute_passive_health_check_alerts.listList passive health check alertsread
Compute regions1
together_ai.compute_regions.listList regions and corresponding supported driver versionsread
Deployments logs1
together_ai.deployments_logs.listGet logs for a deploymentread
Deployments storage files1
together_ai.deployments_storage_files.get_urlGet a file download URLread
Embeddings1
together_ai.embeddings.createCreate embeddingwrite
Evaluation models1
together_ai.evaluation_models.listGet model listread
Evaluation status1
together_ai.evaluation_status.getGet evaluation job status and resultsread
Fine tunes checkpoints1
together_ai.fine_tunes_checkpoints.listList checkpointsread
Fine tunes download tokenized dataset1
together_ai.fine_tunes_download_tokenized_dataset.getDownload tokenized datasetread
Fine tunes events1
together_ai.fine_tunes_events.listList job eventsread
Fine tunes metrics1
together_ai.fine_tunes_metrics.listGet metricsread
Fine tunes models limits1
together_ai.fine_tunes_models_limits.getGet model limitsread
Fine tunes models supported1
together_ai.fine_tunes_models_supported.getList supported modelsread
Image generations1
together_ai.image_generations.createCreate imagewrite
Organizations endpoints1
together_ai.organizations_endpoints.listList organization endpointsread
Organizations models1
together_ai.organizations_models.listList organization modelsread
Projects1
together_ai.projects.listList available projectsread
Projects endpoints analytics1
together_ai.projects_endpoints_analytics.getGet endpoint analyticsread
Projects endpoints events1
together_ai.projects_endpoints_events.listList endpoint eventsread
Projects models files1
together_ai.projects_models_files.listList model filesread
Projects models uploads events1
together_ai.projects_models_uploads_events.listList remote model upload eventsread
Public endpoints1
together_ai.public_endpoints.listList public endpointsread
Queue1
together_ai.queue.clearClear a model's pending jobsdestructive
Queue metrics1
together_ai.queue_metrics.getGet queue metricsread
Rerank1
together_ai.rerank.createCreate a rerank requestwrite
RL checkpoints download urls1
together_ai.rl_checkpoints_download_urls.listDownload checkpointread
RL supported models1
together_ai.rl_supported_models.listList supported modelsread
RL training sessions operations custom forward backward1
together_ai.rl_training_sessions_operations_custom_forward_backward.getGet custom forward-backward operationread
RL training sessions operations forward backward1
together_ai.rl_training_sessions_operations_forward_backward.getGet forward-backward operationread
RL training sessions operations inference checkpoint1
together_ai.rl_training_sessions_operations_inference_checkpoint.getGet inference checkpoint operationread
RL training sessions operations optim step1
together_ai.rl_training_sessions_operations_optim_step.getGet optim-step operationread
RL training sessions operations sample1
together_ai.rl_training_sessions_operations_sample.getGet sample operationread
RL training sessions operations training checkpoint1
together_ai.rl_training_sessions_operations_training_checkpoint.getGet save training checkpoint operationread
RL training sessions operations weights sync1
together_ai.rl_training_sessions_operations_weights_sync.getGet weights-sync operationread
Shaping events1
together_ai.shaping_events.listList shaping job eventsread
Voices1
together_ai.voices.listFetch available voices for each modelread
Whoami1
together_ai.whoami.getIdentify the callerread
Questions
The Together AI MCP server, answered.
Is there a Together AI MCP server for Claude?
Yes. actionlake is an MCP gateway: a remote MCP server that gives Claude, or another MCP client, 183 Together AI actions across projects endpoints rollouts, fine tunes, RL training sessions operations and compute clusters instances remediations, through the same eight tools it uses for every other connector. actionlake is independent; it is not made or endorsed by Together AI.
What do I need to connect Together AI?
Together AI API key (api.together.ai > Settings > API Keys). The free call account { op: "connections.requirements", service: "together_ai" } lists exactly what to paste.
Can an AI agent delete or change things in Together AI?
Only if its key allows it, and only with an explicit confirmation. 25 Together AI actions are destructive, including delete a deployment, delete a file and delete a volume. Each runs only with confirm: true. Under the human confirm policy, or an approval rule on the key, the run waits until a named person approves it.
Will an AI agent use up my Together AI API limit?
actionlake paces every call to the limits in the Together AI service file: 60 calls a minute per connection. Repeated reads are answered from cache (87 Together AI actions can be), which never reaches Together AI.
What does a Together AI action cost?
One upstream call, the only unit actionlake bills. A list that pages bills one call per page (49 Together AI actions page). Cache hits, searches, dry runs and calls Together AI refused with 401, 403, 429 or a 5xx are free. The Free plan includes 100,000 calls a month with no card. Paid plans start at US$49 a month for 250,000 calls, with overage from US$0.30 down to US$0.18 per 1,000.
Which Together AI API does it cover?
The Together AI API 2.0.0 at api.together.ai. It is built from Together AI's documentation as read on 23 October 2026. 183 actions in 74 areas, each one in the signed service file (version 1.0.0).
Does actionlake keep my Together AI data, and where?
actionlake processes and stores data in Sydney, Australia; there is no US region yet. Reads may be cached for up to 5 minutes, and a write clears the reads it affects. Each run's request and response is kept, sealed, for your plan's trace window: 1 day on Free, 3 days on Starter, 7 days on Pro, 14 days on Team, 30 days on Business and 30 days on Enterprise. Your Together AI credential is sealed under your tenant's key, never logged and never returned. Your traffic is not used to train models.
Which AI clients can use it?
Claude Code, tested by us end to end, and the Claude app as a custom connector, tested by us with an existing key. Other MCP clients that support remote servers over Streamable HTTP (MCP revisions 2025-06-18 and 2025-11-25) and OAuth dynamic client registration, or a bearer header, should connect the same way; we have not tested them. Or plain REST, under the same action names.
Give your agent Together AI today.
Start free - 100,000 calls/month, no card. No signup. Hard caps on every key. Cached answers are free.