Serveur MCP
compute.pangle.online — GPU rental market
online.pangle.compute/gpu-rental-market
Ce que fait ce MCP
Searches and analyzes live GPU rental offers and supports watches, limit orders, rental placement, instance status, usage tracking, and destruction.
Capacités
Outils
25
outils
account_usage
Your metered usage ledger: placements, destroys, estimated hours and USD,
and the platform fee (2.5%, $0.01 min, $5.00 cap per lease per UTC day) priced
at the rate live at event time. Distinct from fee_bps, the display markup.
Schéma d’entrée
{'type': 'object', 'title': 'account_usageArguments', 'required': ['account_token'], 'properties': {'days': {'type': 'integer', 'title': 'Days', 'default': 90}, 'account_token': {'type': 'string', 'title': 'Account Token'}}}
best_value
Perf-per-dollar board: for each GPU model, best live price vs its FP16
tensor throughput (dense, vendor spec sheets) -> TFLOPS per dollar-hour.
The answer to 'most compute for my budget' rather than 'cheapest card'.
Models without a defensible public spec figure are listed as unrated.
Schéma d’entrée
{'type': 'object', 'title': 'best_valueArguments', 'properties': {'limit': {'type': 'integer', 'title': 'Limit', 'default': 25}, 'min_vram_gb': {'type': 'number', 'title': 'Min Vram Gb', 'default': 0}, 'offer_class': {'type': 'string', 'title': 'Offer Class', 'default': ''}}}
cancel_limit_order
Cancel a limit order. Requires the order_secret from create_limit_order.
Schéma d’entrée
{'type': 'object', 'title': 'cancel_limit_orderArguments', 'required': ['order_id', 'order_secret'], 'properties': {'order_id': {'type': 'string', 'title': 'Order Id'}, 'order_secret': {'type': 'string', 'title': 'Order Secret'}}}
create_limit_order
Place a standing LIMIT ORDER: when the best live fee-adjusted price for
gpu (and optional offer_class) trades at or under max_price_per_gpu_hr,
the station cuts a signed FILL TICKET naming the exact offer. The station
never holds your provider key â the ticket is executed by whoever does:
the open-source keyholder sidecar (/agents/#sidecar), your own agent
long-polling POST /api/orders/{id}/ticket, or a human with curl. Returns
the order with order_secret SHOWN ONCE â it authenticates ticket reads and
the fill call for this order only and can rent nothing by itself. Triggers
are checked on the poll cadence (~5-min bars, not tick-by-tick); orders
expire in 30 days; a ticket lasts ~4 minutes then the order re-arms.
standing=true makes it a STANDING order â compute that survives: after a
live fill the sidecar keeps watching the machine, and when it is preempted
or dies the order re-arms and refills from the CURRENT best offer, up to
max_refills times with cooloff_seconds between death and refill. Honest
limits: this re-provisions the MACHINE, not the WORK (GPU memory is not
portable â resume from your own checkpoints); refills need your sidecar
alive (it is both witness and executor); cancelling stops supervision but
cannot destroy a running machine (the station holds no key). Watchdog:
spend_alert_usd / age_alert_hours fire a "still yours?" webhook once per
machine when the spend ESTIMATE (uptime x price; the provider's bill is
authoritative) or age crosses your line â auto_destroy_budget_usd remains
the hard stop.
Schéma d’entrée
{'type': 'object', 'title': 'create_limit_orderArguments', 'required': ['gpu', 'max_price_per_gpu_hr'], 'properties': {'gpu': {'type': 'string', 'title': 'Gpu'}, 'standing': {'type': 'boolean', 'title': 'Standing', 'default': False}, 'max_refills': {'type': 'integer', 'title': 'Max Refills', 'default': 5}, 'min_vram_gb': {'type': 'number', 'title': 'Min Vram Gb', 'default': 0}, 'offer_class': {'type': 'string', 'title': 'Offer Class', 'default': ''}, 'webhook_url': {'type': 'string', 'title': 'Webhook Url', 'default': ''}, 'min_gpu_count': {'type': 'integer', 'title': 'Min Gpu Count', 'default': 0}, 'age_alert_hours': {'type': 'number', 'title': 'Age Alert Hours', 'default': 0}, 'cooloff_seconds': {'type': 'integer', 'title': 'Cooloff Seconds', 'default': 120}, 'spend_alert_usd': {'type': 'number', 'title': 'Spend Alert Usd', 'default': 0}, 'max_price_per_gpu_hr': {'type': 'number', 'title': 'Max Price Per Gpu Hr'}, 'auto_destroy_budget_usd': {'type': 'number', 'title': 'Auto Destroy Budget Usd', 'default': 0}}}
create_watch
Create a price watch (tripwire): fires when the best live fee-adjusted
price per GPU-hour matching gpu (and optional offer_class) drops to or
under max_price_per_gpu_hr. Checked every poll (~5 min). Returns the watch
with its id â the id is the only key; poll it with watch_status, or give a
public webhook_url to be POSTed on each trip. Watches re-arm when the
price climbs 2% back over the line.
Schéma d’entrée
{'type': 'object', 'title': 'create_watchArguments', 'required': ['gpu', 'max_price_per_gpu_hr'], 'properties': {'gpu': {'type': 'string', 'title': 'Gpu'}, 'offer_class': {'type': 'string', 'title': 'Offer Class', 'default': ''}, 'webhook_url': {'type': 'string', 'title': 'Webhook Url', 'default': ''}, 'max_price_per_gpu_hr': {'type': 'number', 'title': 'Max Price Per Gpu Hr'}}}
delete_watch
Delete a watch by id.
Schéma d’entrée
{'type': 'object', 'title': 'delete_watchArguments', 'required': ['watch_id'], 'properties': {'watch_id': {'type': 'string', 'title': 'Watch Id'}}}
destroy_rental
Destroy an instance previously placed on your account (vast or runpod).
Your key passes through and is never stored.
Schéma d’entrée
{'type': 'object', 'title': 'destroy_rentalArguments', 'required': ['provider', 'provider_instance_id', 'api_key'], 'properties': {'api_key': {'type': 'string', 'title': 'Api Key'}, 'provider': {'type': 'string', 'title': 'Provider'}, 'provider_instance_id': {'type': 'string', 'title': 'Provider Instance Id'}}}
fill_limit_order
Execute a limit order's live fill ticket on YOUR provider key (used for
this one call, never stored). Only works while the order is ticketed; the
offer is re-quoted live and refused above the order's line. Idempotent per
order â a retry returns the first receipt rather than renting a second
machine. The order's auto_destroy_budget_usd (if set) arms the budget
guard exactly as rent_offer does.
Schéma d’entrée
{'type': 'object', 'title': 'fill_limit_orderArguments', 'required': ['order_id', 'order_secret', 'api_key'], 'properties': {'image': {'type': 'string', 'title': 'Image', 'default': ''}, 'api_key': {'type': 'string', 'title': 'Api Key'}, 'confirm': {'type': 'boolean', 'title': 'Confirm', 'default': False}, 'disk_gb': {'type': 'number', 'title': 'Disk Gb', 'default': 10}, 'dry_run': {'type': 'boolean', 'title': 'Dry Run', 'default': True}, 'order_id': {'type': 'string', 'title': 'Order Id'}, 'order_secret': {'type': 'string', 'title': 'Order Secret'}, 'account_token': {'type': 'string', 'title': 'Account Token', 'default': ''}}}
get_offer
Fetch one offer by its id (from search_offers).
Schéma d’entrée
{'type': 'object', 'title': 'get_offerArguments', 'required': ['offer_id'], 'properties': {'offer_id': {'type': 'string', 'title': 'Offer Id'}}}
idle_history
The idle-capacity index over time: share of listed GPU capacity that is
interruptible (spot/bid) â idle hardware looking for work. Points are
[ts, idle_share, idle_offers, total_offers].
Schéma d’entrée
{'type': 'object', 'title': 'idle_historyArguments', 'properties': {'hours': {'type': 'integer', 'title': 'Hours', 'default': 168}}}
limit_order_status
State of a limit order (armed/ticketed/filled/cancelled/expired), last
price seen, recent events. Pass order_secret to also read the live fill
ticket when one is cut.
Schéma d’entrée
{'type': 'object', 'title': 'limit_order_statusArguments', 'required': ['order_id'], 'properties': {'order_id': {'type': 'string', 'title': 'Order Id'}, 'order_secret': {'type': 'string', 'title': 'Order Secret', 'default': ''}}}
market_stats
Live market snapshot: total offers, GPU models, idle-capacity share,
per-provider freshness/health, cheapest price per model, current fee_bps.
Schéma d’entrée
{'type': 'object', 'title': 'market_statsArguments', 'properties': {}}
price_history
Price history for one GPU model (exact name from search_offers/market_stats,
e.g. 'RTX 4090'). offer_class: on_demand | interruptible. Returns per-provider
series of [ts, min_price_per_gpu_hr, median_price_per_gpu_hr, offer_count];
ranges past 3 days are bucketed (hourly, then 6-hourly past a week).
Schéma d’entrée
{'type': 'object', 'title': 'price_historyArguments', 'properties': {'gpu': {'type': 'string', 'title': 'Gpu', 'default': ''}, 'hours': {'type': 'integer', 'title': 'Hours', 'default': 168}, 'gpu_model': {'type': 'string', 'title': 'Gpu Model', 'default': ''}, 'offer_class': {'type': 'string', 'title': 'Offer Class', 'default': 'on_demand'}}}
price_position
Rent-now-or-wait context: where the current best price for one GPU
model sits inside its own trailing 7/30-day range (percentile, 24h trend,
verdict). Descriptive, not a forecast.
Schéma d’entrée
{'type': 'object', 'title': 'price_positionArguments', 'properties': {'gpu': {'type': 'string', 'title': 'Gpu', 'default': ''}, 'gpu_model': {'type': 'string', 'title': 'Gpu Model', 'default': ''}, 'offer_class': {'type': 'string', 'title': 'Offer Class', 'default': 'on_demand'}}}
provider_reliability
How dependable each provider's data feed has been for this station
(poll success over the trailing window), plus the provider's own average
machine reliability where reported. feed_score is about the DATA, not
their hardware.
Schéma d’entrée
{'type': 'object', 'title': 'provider_reliabilityArguments', 'properties': {'days': {'type': 'integer', 'title': 'Days', 'default': 30}}}
register_account
Create a metering account: returns a bearer token (shown once, we keep
only a hash). Pass it as account_token on rent calls to build a usage
ledger you can read back with account_usage. Optional â renting works
without one.
Schéma d’entrée
{'type': 'object', 'title': 'register_accountArguments', 'properties': {'label': {'type': 'string', 'title': 'Label', 'default': ''}}}
rental_status
Live state of an instance on your account: running/starting/stopped/
gone, uptime, price, estimated spend so far, and the budget guard's state
if one is armed. Read-only; your key passes through and is never stored.
Schéma d’entrée
{'type': 'object', 'title': 'rental_statusArguments', 'required': ['provider', 'provider_instance_id', 'api_key'], 'properties': {'api_key': {'type': 'string', 'title': 'Api Key'}, 'provider': {'type': 'string', 'title': 'Provider'}, 'provider_instance_id': {'type': 'string', 'title': 'Provider Instance Id'}}}
rent_best
One-shot rent: search the live book with these filters and place on the
best (cheapest) matching offer â 'cheapest H100 under $2/hr, go'. Same
guarantees as rent_offer (dry-run default, live re-quote, your
max_price_per_gpu_hr is absolute). If the best offer is already gone it
walks to the next, at most 3, and reports what it skipped.
Schéma d’entrée
{'type': 'object', 'title': 'rent_bestArguments', 'required': ['api_key', 'max_price_per_gpu_hr'], 'properties': {'gpu': {'type': 'string', 'title': 'Gpu', 'default': ''}, 'image': {'type': 'string', 'title': 'Image', 'default': ''}, 'label': {'type': 'string', 'title': 'Label', 'default': ''}, 'api_key': {'type': 'string', 'title': 'Api Key'}, 'confirm': {'type': 'boolean', 'title': 'Confirm', 'default': False}, 'country': {'type': 'string', 'title': 'Country', 'default': ''}, 'disk_gb': {'type': 'number', 'title': 'Disk Gb', 'default': 10}, 'dry_run': {'type': 'boolean', 'title': 'Dry Run', 'default': True}, 'provider': {'type': 'string', 'title': 'Provider', 'default': ''}, 'min_vram_gb': {'type': 'number', 'title': 'Min Vram Gb', 'default': 0}, 'offer_class': {'type': 'string', 'title': 'Offer Class', 'default': ''}, 'account_token': {'type': 'string', 'title': 'Account Token', 'default': ''}, 'min_gpu_count': {'type': 'integer', 'title': 'Min Gpu Count', 'default': 0}, 'idempotency_key': {'type': 'string', 'title': 'Idempotency Key', 'default': ''}, 'max_price_per_gpu_hr': {'type': 'number', 'title': 'Max Price Per Gpu Hr'}, 'auto_destroy_budget_usd': {'type': 'number', 'title': 'Auto Destroy Budget Usd', 'default': 0}}}
rent_offer
Place a rental on YOUR provider account (BYO key â it is used for this
one call and never stored). Default is a DRY RUN returning exactly what
would be executed; a live placement needs dry_run=false AND confirm=true.
The offer is re-quoted live first and refused if its price exceeds
max_price_per_gpu_hr. Executes on vast and runpod offers. Pass an
idempotency_key so retries return the first receipt instead of renting
twice. auto_destroy_budget_usd arms the budget guard: this station then
holds your key in PROCESS MEMORY ONLY and destroys the rental when
estimated spend reaches the cap (a station restart drops the guard â the
receipt says so; poll rental_status as backstop). account_token (from
register_account) attributes the rental to your usage ledger.
Schéma d’entrée
{'type': 'object', 'title': 'rent_offerArguments', 'required': ['offer_id', 'api_key', 'max_price_per_gpu_hr'], 'properties': {'image': {'type': 'string', 'title': 'Image', 'default': ''}, 'label': {'type': 'string', 'title': 'Label', 'default': ''}, 'api_key': {'type': 'string', 'title': 'Api Key'}, 'confirm': {'type': 'boolean', 'title': 'Confirm', 'default': False}, 'disk_gb': {'type': 'number', 'title': 'Disk Gb', 'default': 10}, 'dry_run': {'type': 'boolean', 'title': 'Dry Run', 'default': True}, 'offer_id': {'type': 'string', 'title': 'Offer Id'}, 'account_token': {'type': 'string', 'title': 'Account Token', 'default': ''}, 'idempotency_key': {'type': 'string', 'title': 'Idempotency Key', 'default': ''}, 'max_price_per_gpu_hr': {'type': 'number', 'title': 'Max Price Per Gpu Hr'}, 'auto_destroy_budget_usd': {'type': 'number', 'title': 'Auto Destroy Budget Usd', 'default': 0}}}
rent_x402
Rent a prepaid block of GPU time with a wallet and NO account, NO API key
and no human, over x402 (HTTP 402 micropayments, USDC on Base).
Call it with no arguments and you get the payment challenge: an x402
PaymentRequirementsResponse whose accepts[0].extra.x402_fee states the block
price, the platform fee IN DOLLARS, that there are no refunds, and that you
are charged only after the lease is live. Sign the EIP-3009
transferWithAuthorization it describes, then call again with x_payment set to
the base64 PaymentPayload (the same value the HTTP rail sends as X-PAYMENT).
`ssh_pubkey` is HOW YOU GET IN, and is required on the paying call: pass the
public half of your ssh key (one line, the contents of id_ed25519.pub). It is
written into the rented container's authorized_keys and nowhere else, and the
receipt comes back with the host and port to connect to. A block nobody can
enter is not sold.
`lease_id` is minted by the challenge â send back the one you were given; it
is the accrual key for the daily fee cap and this station refuses ids it did
not mint or that belong to another payer.
`originator` is an OPTIONAL opaque attribution label (<=64 chars, e.g. a
handle or a reverse-DNS name) naming the client implementation that produced
the fill. It is carried unmodified onto the receipt and onto the public
ledger row, and it never affects the price, the fee, verification or
settlement â the same block costs the same either way.
Returns the receipt on success. Every fill is published at /api/receipts.
Wire contract: docs/X402-FEE-0.1.md, frozen as x402-fee/0.1.
Note for x402 MCP-transport clients: this tool takes the payment as an
ARGUMENT rather than in _meta["x402/payment"]. The payload is byte-identical;
only the channel differs, because a tool argument is the one place every MCP
client can portably put it. The HTTP rail is fully transport-conformant.
Schéma d’entrée
{'type': 'object', 'title': 'rent_x402Arguments', 'properties': {'blocks': {'type': 'integer', 'title': 'Blocks', 'default': 1}, 'lease_id': {'type': 'string', 'title': 'Lease Id', 'default': ''}, 'x_payment': {'type': 'string', 'title': 'X Payment', 'default': ''}, 'originator': {'type': 'string', 'title': 'Originator', 'default': ''}, 'ssh_pubkey': {'type': 'string', 'title': 'Ssh Pubkey', 'default': ''}}}
search_offers
Search live GPU offers. gpu matches the model name (e.g. '4090', 'H100').
offer_class: on_demand | interruptible | reserved. country: ISO-2 (e.g. US),
a comma list ('DE,NL'), or 'EU' for the 27 member states. min_gpu_count
filters to multi-GPU configs (8 = full 8x nodes). region substring-matches
the provider's region/datacenter label. Returns offers ranked
cheapest-first by price_per_gpu_hr.
Schéma d’entrée
{'type': 'object', 'title': 'search_offersArguments', 'properties': {'gpu': {'type': 'string', 'title': 'Gpu', 'default': ''}, 'limit': {'type': 'integer', 'title': 'Limit', 'default': 20}, 'region': {'type': 'string', 'title': 'Region', 'default': ''}, 'country': {'type': 'string', 'title': 'Country', 'default': ''}, 'provider': {'type': 'string', 'title': 'Provider', 'default': ''}, 'min_vram_gb': {'type': 'number', 'title': 'Min Vram Gb', 'default': 0}, 'offer_class': {'type': 'string', 'title': 'Offer Class', 'default': ''}, 'min_gpu_count': {'type': 'integer', 'title': 'Min Gpu Count', 'default': 0}, 'min_reliability': {'type': 'number', 'title': 'Min Reliability', 'default': 0}, 'max_price_per_gpu_hr': {'type': 'number', 'title': 'Max Price Per Gpu Hr', 'default': 0}}}
Schéma de sortie
{'type': 'object', 'title': 'search_offersOutput', 'required': ['result'], 'properties': {'result': {'type': 'array', 'items': {'type': 'object', 'additionalProperties': True}, 'title': 'Result'}}}
spot_spread
On-demand vs interruptible spread: the live discount for idle capacity
per model; pass gpu_model for its history too.
Schéma d’entrée
{'type': 'object', 'title': 'spot_spreadArguments', 'properties': {'hours': {'type': 'integer', 'title': 'Hours', 'default': 168}, 'gpu_model': {'type': 'string', 'title': 'Gpu Model', 'default': ''}}}
true_cost_notes
The billing fine print behind the $/hr, verified per provider from
their own published pages (source URL + date on every claim): billing
granularity (per-second vs per-minute), whether storage bills while
stopped, bandwidth/egress fees, and the commonly-missed gotchas (e.g.
Vast bandwidth never appears in $/hr; RunPod stopped volume disk costs
double). Providers we have not verified are listed with NO claims.
Schéma d’entrée
{'type': 'object', 'title': 'true_cost_notesArguments', 'properties': {}}
watch_status
Current state of a watch: armed/tripped, last price seen, recent
events, and the best matching offer right now.
Schéma d’entrée
{'type': 'object', 'title': 'watch_statusArguments', 'required': ['watch_id'], 'properties': {'watch_id': {'type': 'string', 'title': 'Watch Id'}}}
will_it_fit
Will this model fit, and what does a fitting machine cost right now?
Deterministic VRAM arithmetic â weights (params x bytes/precision) + KV
cache (2 x layers x kv_heads x head_dim x context x batch) + stated
overhead â checked against the live book, cheapest fitting offers first.
Pass a known model slug (llama-3.1-8b/70b/405b, mistral-7b, mixtral-8x7b,
qwen2.5-7b/32b/72b, gemma-2-9b/27b) OR params_b for a stated Llama-like
guess. precision: fp16/int8/int4. If you pass tokens_per_s (YOUR measured
throughput â we never invent one), each offer gains usd_per_m_tokens.
Every approximation is in the assumptions list, not fine print.
Schéma d’entrée
{'type': 'object', 'title': 'will_it_fitArguments', 'properties': {'batch': {'type': 'integer', 'title': 'Batch', 'default': 1}, 'limit': {'type': 'integer', 'title': 'Limit', 'default': 10}, 'model': {'type': 'string', 'title': 'Model', 'default': ''}, 'context': {'type': 'integer', 'title': 'Context', 'default': 8192}, 'params_b': {'type': 'number', 'title': 'Params B', 'default': 0}, 'precision': {'type': 'string', 'title': 'Precision', 'default': 'fp16'}, 'offer_class': {'type': 'string', 'title': 'Offer Class', 'default': ''}, 'tokens_per_s': {'type': 'number', 'title': 'Tokens Per S', 'default': 0}}}
Historique
Modifications récentes des outils
Ajouté
fill_limit_order
17 September 2026 12:54
Ajouté
destroy_rental
17 September 2026 12:54
Ajouté
rental_status
17 September 2026 12:54
Ajouté
rent_best
17 September 2026 12:54
Ajouté
rent_offer
17 September 2026 12:54
Ajouté
rent_x402
17 September 2026 12:54
Ajouté
cancel_limit_order
17 September 2026 12:54
Ajouté
limit_order_status
17 September 2026 12:54
Ajouté
create_limit_order
17 September 2026 12:54
Ajouté
delete_watch
17 September 2026 12:54
Ajouté
watch_status
17 September 2026 12:54
Ajouté
create_watch
17 September 2026 12:54
Ajouté
account_usage
17 September 2026 12:54
Ajouté
register_account
17 September 2026 12:54
Ajouté
true_cost_notes
17 September 2026 12:54
Ajouté
will_it_fit
17 September 2026 12:54
Ajouté
spot_spread
17 September 2026 12:54
Ajouté
price_position
17 September 2026 12:54
Ajouté
provider_reliability
17 September 2026 12:54
Ajouté
best_value
17 September 2026 12:54
Ajouté
idle_history
17 September 2026 12:54
Ajouté
price_history
17 September 2026 12:54
Ajouté
market_stats
17 September 2026 12:54
Ajouté
get_offer
17 September 2026 12:54
Ajouté
search_offers
17 September 2026 12:54
Découvrir
Serveurs MCP similaires
Conexcol — catálogo de hosting y servidores cloud en Colombia
co.net.conexcol.mcp/hosting-cloud-colombia
Provides a current catalog of Colombian hosting and cloud-server plans, infrastructure facts, pricing, quotations, draft orders, …
Cheapest Domains
domains.cheapest/cheapest-domains
Checks domain availability, compares registrar registration and renewal prices, estimates costs, and provides registrar search li…
MCPDomain
io.github.danboabes/mcpdomain
Checks domain availability, registers and transfers domains, and manages DNS, email forwarding, and domain status.
TitanStore Commerce MCP
io.github.Fish-Tacos/titan-store
Searches and sells compute credits, API access, GPU instances, cloud storage, and support plans.
GrabGPU
com.grabgpu/gpu-finder
Compares cloud GPU rental offers, hourly prices, reported stock, providers, and rental links.
BorealHost
ai.borealhost/mcp
Provides web hosting and infrastructure management, including site deployment, DNS, domains, containers, compute, backups, cachin…
AgentLot Marketplace
io.agentlot/marketplace
Enables agents and users to discover, bid on, purchase, deliver, and settle paid digital services and compute work.
apex-x1
xyz.apexfaucet/apex-x1
Provides X1 and Arc token screening, liquidity and exit-risk checks, wallet and bridge data, non-custodial token operations, comm…