# Thunder Compute

> Thunder Compute is a US GPU cloud renting RTX A6000, L40, A100 and H100 instances by the minute and GPU sandboxes by the second. Agents use its REST API, a hosted MCP server with OAuth, or the tnr CLI.

- Canonical: https://www.anchorterminal.com/tools/thunder-compute
- Markdown: https://www.anchorterminal.com/tools/thunder-compute.md (~8,550 tokens)
- Slim: https://www.anchorterminal.com/tools/thunder-compute.min.md (~1,880 tokens, same facts, less prose, for token-sensitive contexts)
- JSON: https://www.anchorterminal.com/tools/thunder-compute.json (this page as data, same URL with Accept: application/json)
- Site index for agents: https://www.anchorterminal.com/llms.txt (full text: https://www.anchorterminal.com/llms-full.txt)
- API: https://www.anchorterminal.com/api/v1/index.json
- Updated: 2026-10-09

## Overview

**Grade C · 56.1/100 · rank #576 of 842 · #10 in GPU & serverless compute · not agent-ready · confidence medium**


## Assessment

The hosted MCP server signs in with OAuth and separate read and write scopes, and the REST API has a public OpenAPI 3.1 document with keyless price and availability endpoints. No rate limits, SLA or API changelog were found, instance creation has no idempotency key, and instances cannot be stopped, only deleted.

## Facts

| Field | Value |
| --- | --- |
| Vendor | Thunder Compute (https://www.thundercompute.com) |
| Kind | HTTP API |
| Category | GPU & serverless compute (https://www.anchorterminal.com/categories/gpu-compute) |
| Transport | HTTP, Streamable HTTP |
| Endpoint | `https://api.thundercompute.com:8443/v1` |
| Auth | OAuth or key · Self-serve. The hosted MCP server uses OAuth 2.0 with discovery. The first connection opens a browser to sign in and approve, tokens refresh, and the scopes are `instances:read`, `instances:write`, `snapshots:read`, `snapshots:write`, `sandboxes:read`, `sandboxes:write`, `billing:read` and `org:read`. The REST API takes a named API token as a Bearer header. Tokens are created and revoked one at a time in the console, or with the MCP `create_token` tool, and no token scopes were found. The CLI and the sandbox SDK read `TNR_API_TOKEN` for headless use. Price, specification, availability and template endpoints need no token. |
| Pricing | Pay per use ($0.04 / vCPU-hr) · No general free tier. Instances are billed by the minute while they run, paid by saved card or preloaded credit that is non-refundable. Per GPU an hour on 9 October 2026, RTX A6000 is $0.35, L40 $0.79, A100 80 GB $1.09 and H100 $3.20. Sandboxes are billed by the second and need access enabled by the vendor. Students with a US university email get $20 of credit for A6000 and L40 GPUs. Reserved capacity is by quote (https://www.thundercompute.com/pricing). |
| x402 | No · No x402, MPP or L402 in the docs corpus, the OpenAPI document or the pricing page (checked 2026-10-09). |
| Licence | Proprietary service under Thunder Compute's Terms and Conditions. The `tnr` CLI on GitHub is MIT |
| Tools exposed | 28 |
| Packages | go: `github.com/Thunder-Compute/thunder-cli`; pypi: `thunder-sandbox` |
| MCP registry name | `io.github.Thunder-Compute/thunder-compute` |
| Source | https://github.com/Thunder-Compute/thunder-cli |
| Docs | https://www.thundercompute.com/docs |
| llms.txt | https://www.thundercompute.com/docs/llms.txt |
| Last release | 2026-09-16 |
| GitHub stars | 34 (as of 2026-10-09) |
| Surface graded | The REST API at https://api.thundercompute.com:8443/v1 (OpenAPI 3.1, 41 operations) together with the hosted MCP server, whose tool list comes from the docs because the definitions need sign-in |
| MCP server | Hosted, Streamable HTTP, OAuth. 28 tools. `list_instances`, `create_instance`, `delete_instance`, `modify_instance`, `run_command`, `get_specs`, `get_availability`, `get_pricing`, `list_templates`, `list_snapshots`, `create_snapshot`, `delete_snapshot`, `list_ssh_keys`, `create_ssh_key`, `delete_ssh_key`, `add_ssh_key_to_instance`, `list_ports`, `forward_port`, `delete_port`, `get_ssh_command`, `get_scp_command`, `get_meter_data`, `get_upcoming_invoice`, `get_invoice_history`, `get_subscription`, `list_tokens`, `create_token`, `delete_token` |
| Free tier | None in general. $20 of credit for students with a US university email, usable on A6000 and L40 |
| GPUs | RTX A6000 48 GB (one per instance), L40 48 GB, A100 80 GB and H100 80 GB in 1, 2, 4 or 8 GPU shapes, per the pricing page |
| Scale to zero | Not for instances, which bill until deleted and cannot be stopped. Sandboxes end at a timeout and bill by the second |
| Cold start | No figure published. The pricing page says instances are ready to SSH into in minutes |
| Billing basis | Instances per minute while running, sandboxes per second, snapshots and extra disk per GB metered hourly. No egress fee |
| Rate limits | Not found in the reviewed documentation |
| Errors | `code`, `error` and `message`, for example 400 with `invalid_request` |
| Regions | North America per the technical specifications. No region list found |
| SDKs | `tnr` CLI in Go, v2.1.0 (16 September 2026), MIT. `thunder-sandbox` for Python 3.10 or newer, covering sandboxes only. VS Code and Cursor extension |
| Sandboxes | Firecracker microVMs with 3 to 40 vCPUs, an optional H100 or A100, images from a registry or Dockerfile and an outbound network policy. Access is enabled for each organisation on request |
| Preinstalled | CUDA 13.0, driver 580, PyTorch 2.9.0 and JupyterLab, with templates for Ollama, ComfyUI, Forge Neo and Unsloth |
| Status | status.thundercompute.com on incident.io, with Website, API and Instances components |
| Capabilities | compute.gpu, compute.containers |
| Tags | hosted, usage-priced, openapi, llms-txt, mcp, cli, oauth, api-key, python, go, status-page, card-required, closed-source |
| JSON | https://www.anchorterminal.com/api/v1/tools/thunder-compute.json |

## Score breakdown (methodology v0.4, October 2026 research run)

Assessed 2026-10-09 from public evidence against the published checklist (https://www.anchorterminal.com/benchmark/#checklist). Confidence: medium. Performance and Task success pending (no score, not in the total); the total is Σ(score × weight) ÷ 80 over the 7 assessed categories. "This run" is each category's share of the 100 points.

| Category | Weight | This run | Score (0–100) | Points |
| --- | --- | --- | --- | --- |
| Reliability | 16% | 20 | 65 | 13.0 |
| Performance | 10% | pending | pending | n/a |
| Schema & documentation | 13% | 16.2 | 70 | 11.4 |
| Agent ergonomics | 13% | 16.2 | 48 | 7.8 |
| Security & auth | 14% | 17.5 | 51 | 8.9 |
| Payments & pricing | 10% | 12.5 | 20 | 2.5 |
| Task success | 10% | pending | pending | n/a |
| Maintenance & community | 7% | 8.8 | 79 | 6.9 |
| Transparency & trust (editorial 51, provenance 77) | 7% | 8.8 | 64 | 5.6 |
| Negative events | up to −15 | up to −15 | none recorded | 0 |
| **Total** | | | | **56.1 → C** |

### Why each score

- Reliability 65: Graded on the hosted lines for the REST API and the hosted MCP server. Status page on incident.io at status.thundercompute.com with Website, API and Instances components and monthly history (20). No incidents listed for July to October 2026, and each component shows 100 per cent (30). No rate limits with numbers were found in the docs or the OpenAPI document (0). No 429 or retry guidance in the docs. The CLI source reads a `Retry-After` header on errors, adding a duplicate SSH key is a no-op and port updates are atomic, but instance creation has no idempotency key (5 of 15). No SLA found, and the terms say availability is not guaranteed (0). The API and MCP server carry no beta label. Sandboxes need access enabled for each organisation (10).
- Performance: Pending. Latency is measured per call by our probes, which haven't run yet, so this run doesn't score it. Its weight is shared across the assessed categories until the first probe window closes.
- Schema & documentation 70: Public OpenAPI 3.1 document at api.thundercompute.com:8443/openapi.json with 41 operations and 72 schemas (25). `llms.txt`, `llms-full.txt` and a Markdown copy of every docs page (10). 28 of 41 operations have a description. Instance operations state side effects such as the private key being returned once, while most sandbox operations have a summary only (12 of 20). Required fields are marked, but the document has six enums, `gpu_type` and `template` are free strings and the instance list is a map keyed by ID (8 of 15). 12 examples and one error schema with `code`, `error` and `message`. Most operations document 400, 401 and 500, and none documents 429 (9 of 15). `/v1` and `/v2` paths, with legacy endpoints and the `mode` field marked as retained or deprecated. No API changelog was found. The CLI has a dated one (6 of 15). The MCP tool definitions need a signed-in session and were not read.
- Agent ergonomics 48: No field selection, and `GET /instances/list` returns every instance in one map. The MCP server has 28 tools in one set, with no toolsets found (12 of 25). `limit` up to 100, `page_token` and a status filter on sandboxes and sandbox logs. No pagination on instances, snapshots or keys (8 of 20). Errors carry a numeric `code`, a string `error` such as `invalid_request` and a `message`, documented per operation (14 of 20). Duplicate SSH keys are a no-op and port changes are deduplicated in one transaction, but create calls have no idempotency key. MCP tool annotations were not read (7 of 20). A create call needs five fields. The official Python SDK covers sandboxes only, and the Go CLI covers instances (7 of 15).
- Security & auth 51: The MCP server uses OAuth 2.0 with discovery, refresh tokens and ten scopes that separate read from write for instances, snapshots and sandboxes. The REST API uses named API tokens that are created and revoked one at a time, with no scopes found. No secret travels in a URL (25 of 30). Read scopes exist, and the CLI requires `--yes` to delete without a prompt. No confirmation step for `delete_instance` or `delete_snapshot` was found in the MCP docs, and both are described as irreversible (8 of 20). Responses are infrastructure metadata, except `run_command` output and sandbox logs, with no guidance found on untrusted output (8 of 15). `GET /tokens` returns creation and last-used times and sandbox logs are listed by API. No audit log was found (4 of 15). No security.txt, disclosure policy or bug bounty was found. `llms.txt` states SOC 2 Type II and GDPR compliance with no report or trust page linked, and the DPA promises an audit summary on written request and incident notice within 72 hours (6 of 20).
- Payments & pricing 20: No x402, MPP or L402 in the docs, the OpenAPI document or the pricing page (0). Per-GPU-hour, per-vCPU-hour and storage prices are public on the pricing page and from `GET /v2/pricing` without a token (20). No general free tier or trial. A $20 credit is limited to students with a US university email, and the MCP quickstart asks for a payment method before connecting (0). Signup and the first OAuth consent happen in a browser. `create_token` issues API tokens only after that sign-in (0).
- Task success: Pending. Task success needs the category task suites run through each tool, which haven't run yet, so this run doesn't score it. Its weight is shared across the assessed categories until then. A data provider's data-quality score is published on its listing now and becomes half of this category when it's scored.
- Maintenance & community 79: The newest CLI release is v2.1.0 of 16 September 2026, and the terms and privacy policy were updated on 28 September 2026 (30). Four CLI releases in the last 90 days, on 15 July, 30 July, 15 September and 16 September 2026 (20). A dated CLI changelog, a support address and a Discord server the docs say the founders answer. No API changelog, and the CLI repository shows no open issues (10 of 15). Listed in the official MCP registry as `io.github.Thunder-Compute/thunder-compute`, version 1.0.0 from 9 April 2026, with a remote address that differs from the one in the docs. The CLI and the `thunder-sandbox` SDK are current (13 of 15). The CLI repository has tests, a lint configuration and release checksum verification. No public CI workflow was found in it (6 of 10).
- Transparency & trust 64: Closed service under Terms and Conditions dated 28 September 2026 and Delaware law. The CLI is MIT. The OpenAPI document names Apache 2.0 and a placeholder terms address (15 of 30). The privacy policy of 28 September 2026 says instance data is not used to train AI models, the terms say instance data may be deleted 60 days after the account was last active, and a DPA is published. The sub-processor list names Firebase for sign-in while the MCP docs name Stytch, and the FAQ says instances are EC2 while the list names Azure (18 of 30). No deprecation policy. The terms allow changes without notice, and the billing docs say rates can change without notice. Legacy endpoints are marked as retained for compatibility (6 of 20). Five sub-processors with locations are listed and the DPA sets 10 business days' notice of changes. The list is dated 27 August 2025 and omits Stytch and Sentry, which the MCP docs and the CLI source name (12 of 20).

Fix list for a coding agent, everything this grade says the listing lacks, the biggest gain first (20 items): https://www.anchorterminal.com/fixes/thunder-compute.md (JSON https://www.anchorterminal.com/fixes/thunder-compute.json)

### What we couldn't check

- The lead was right about the interfaces. It did not mention sandboxes, which are a second product on the same API and need access enabled for each organisation.
- unchecked: the MCP tool definitions, input schemas and annotations. `tools/list` answers 401 without a signed-in session, so the 28 tools are counted from the docs table.
- unchecked: the PyPI pages for `thunder-sandbox` and `tnr`. The project page needs scripts and PyPI's robots.txt disallows the JSON API, so versions, dates and downloads were not read.
- unchecked: the SOC 2 Type II and GDPR claims in `llms.txt`. No report, auditor or trust page was found on the site.
- The docs give the MCP address as `https://www.thundercompute.com/mcp`. The official registry entry of 9 April 2026 gives `https://api.thundercompute.com:8443/mcp`. We did not test the second.
- `llms.txt` has sections titled Preferred Summary Style for AI and Additional Guidance for AI that tell models how to describe the company. Recorded as a fact. We did not act on them and took no deduction.
- The CLI source initialises Sentry error reporting when a DSN is built in. No opt-out or mention was found in the docs, and Sentry is not on the sub-processor list. The CLI is not the surface graded, so no deduction was taken.
- The FAQ says instances are EC2 instances, the sub-processor list names Azure, and the technical specifications say only North America. The hosting provider and regions are not established.
- `llms.txt` describes prototyping and production modes. The OpenAPI document marks `mode` as deprecated and the pricing page no longer shows modes.
- Whether a user can grant only read scopes at the OAuth consent screen is unknown.

### Sources

- site llms.txt, with SOC 2 claim and guidance addressed to AI models: <https://www.thundercompute.com/llms.txt> (seen 2026-10-09)
- docs index for agents: <https://www.thundercompute.com/docs/llms.txt> (seen 2026-10-09)
- whole docs corpus (MCP server, billing, restrictions, sandboxes, DPA, stopping instances): <https://www.thundercompute.com/docs/llms-full.txt> (seen 2026-10-09)
- MCP server guide with tools, prompts and setup: <https://www.thundercompute.com/docs/guides/mcp-server.md> (seen 2026-10-09)
- MCP quickstart: <https://www.thundercompute.com/docs/mcp-server/quickstart.md> (seen 2026-10-09)
- MCP OAuth protected resource metadata with scopes: <https://www.thundercompute.com/.well-known/oauth-protected-resource> (seen 2026-10-09)
- OpenAPI document: <https://api.thundercompute.com:8443/openapi.json> (seen 2026-10-09)
- public price endpoint: <https://api.thundercompute.com:8443/v2/pricing> (seen 2026-10-09)
- CLI authentication and API tokens: <https://www.thundercompute.com/docs/cli/operations/authentication.md> (seen 2026-10-09)
- pricing: <https://www.thundercompute.com/pricing> (seen 2026-10-09)
- student credit: <https://www.thundercompute.com/students> (seen 2026-10-09)
- billing docs: <https://www.thundercompute.com/docs/billing.md> (seen 2026-10-09)
- Terms and Conditions: <https://www.thundercompute.com/terms-and-conditions> (seen 2026-10-09)
- privacy policy: <https://www.thundercompute.com/privacy-policy> (seen 2026-10-09)
- sub-processors: <https://www.thundercompute.com/sub-processors> (seen 2026-10-09)
- Data Processing Addendum: <https://www.thundercompute.com/docs/guides/data-processing-addendum.md> (seen 2026-10-09)
- status page: <https://status.thundercompute.com/> (seen 2026-10-09)
- status history: <https://status.thundercompute.com/history> (seen 2026-10-09)
- security.txt, answered 404: <https://www.thundercompute.com/.well-known/security.txt> (seen 2026-10-09)
- CLI repository, changelog, tags and client source: <https://github.com/Thunder-Compute/thunder-cli> (seen 2026-10-09)
- official MCP registry entry: <https://registry.modelcontextprotocol.io/v0/servers?search=thunder> (seen 2026-10-09)
- RDAP for thundercompute.com: <https://rdap.verisign.com/com/v1/domain/thundercompute.com> (seen 2026-10-09)

## Who's behind it (provenance 77/100, checked 2026-10-09)

| Check | Finding | Points |
| --- | --- | --- |
| Legal entity named | Thunder GPU, Inc. | 20/20 |
| Domain age | thundercompute.com, registered 2024-02-29 (2 years) | 7/15 |
| Endpoint on the vendor's domain | api.thundercompute.com:8443 | 15/15 |
| Terms of service | read, states 6 of the 7 things a reader expects, and has 2 clauses that cost points | 5.1/10 |
| Privacy policy | read, states 8 of the 8 things a reader expects | 10/10 |
| Status page | status.thundercompute.com | 10/10 |
| Changelog | published | 10/10 |
| security.txt | not found | 0/10 |

The Terms and Conditions, last updated 28 September 2026, name Thunder GPU, Inc., doing business as Thunder Compute, a Delaware company, and apply Delaware law. They cover the site and the services, with sections on automated access, usage billing and instance data deletion.

The privacy policy, last updated 28 September 2026, covers the website and the GPU platform, including access by AI agents.

The API answers at api.thundercompute.com on port 8443, and the MCP server at www.thundercompute.com/mcp, both on the vendor domain.

/.well-known/security.txt answers 404.

RDAP for thundercompute.com gives a registration date of 2024-02-29.

The changelog linked is the CLI's. No API changelog was found.

A Data Processing Addendum is published in the docs and a sub-processor list on the site, last updated 27 August 2025.

### Terms and privacy, as read

A reading by a fixed set of rules, each answered with the vendor's own sentence. Not legal advice.

**Terms of service** (https://www.thundercompute.com/terms-and-conditions), read 2026-10-08, dated 2026-09-28, states 6 of the 7 things a reader expects.

- To know. Restricts benchmarking or competitive use (costs points). "Use the Services as part of any effort to compete with us or otherwise use the Services and/or the Content for any revenue-generating endeavor or commercial enterprise."
- To know. Says the terms or the service can change without notice (costs points). "We reserve the right to change, modify, or remove the contents of the Services at any time or for any reason at our sole discretion without notice."
- To know. Says access can be ended without notice or for any reason. "WE MAY TERMINATE YOUR USE OR PARTICIPATION IN THE SERVICES OR DELETE YOUR ACCOUNT AND ANY CONTENT OR INFORMATION THAT YOU POSTED AT ANY TIME, WITHOUT WARNING, IN OUR SOLE DISCRETION."
- To know. Requires arbitration or waives class actions. "If the Parties are unable to resolve a Dispute through informal negotiations, the Dispute (except those Disputes expressly excluded below) will be finally and exclusively resolved by binding arbitration."
- Gives the date it was last updated. Last updated 2026-09-28.
- Names the governing law or courts. The law of the State of Delaware.
- States a limit on its liability. Capped at the fees paid in the 6 months before the claim.
- Says how changes to the terms are announced. Says it gives notice of a change.
- Not found in the text. Refers to a service level or uptime commitment.
- Also in the text (2026-10-08). Section 28 permits access by AI agents, bots, scripts, command-line tools, SDKs and APIs through the customer's own account, and says it controls where other sections conflict. "you may access and use the Services through automated or non-human means, including AI agents, bots, scripts, command-line tools, SDKs, and our APIs, including to provision, manage, and pay for Services"
- Also in the text (2026-10-08). The customer is responsible for every action an automated system or agent takes on its behalf, as if the customer had taken it. "You are responsible for all actions taken by any automated system or agent operating on your behalf as if you had taken them yourself."
- Also in the text (2026-10-08). Instance data, including storage volumes and snapshots, may be deleted 60 days after the account was last active. "Instance data, including storage volumes, snapshots, and any relevant account data, may be deleted 60 days after account was last active."

**Privacy policy** (https://www.thundercompute.com/privacy-policy), read 2026-10-08, dated 2026-09-28, states 8 of the 8 things a reader expects.

- Gives the date it was last updated. Last updated 2026-09-28.
- Says how long data is kept. For as long as needed, with no period named.
- Says whether personal data is sold or shared for advertising. Says it does not sell personal data.
- Gives a privacy contact. support@thundercompute.com.
- Says where data is transferred or stored. Relies on standard contractual clauses.
- Also in the text (2026-10-08). The policy says content stored or processed on instances is not used to train AI models and is accessed only to run and support the service, prevent abuse or meet legal requirements. "We do not access, inspect, or use Instance Data except as needed to provide and support the Services at your request, to protect the Services against abuse, fraud, or security threats, or as required by law, and we do not use Instance Data to train AI models."
- Also in the text (2026-10-08). Activity by an AI agent, script or other automated tool is processed in the same way as the customer's own activity and is tied to the customer's account. "If you use an AI agent, script, command-line tool, or other automated means to access the Services on your behalf, we process the resulting activity (such as API requests and instance actions) in the same way as activity you perform directly, and it is associated with your account."

## Live (updated 2026-10-09 10:14 UTC)

- Right now: up, HTTP 404, 253 ms, checked 2026-10-09 10:14 UTC (get on `https://api.thundercompute.com:8443/v1`)
- Uptime 24h 100.0% (28 probes) · 30 days 100.0% (28 probes) · p50 618 ms · p95 730 ms
- Vendor status page: none, All Systems Operational
- Always current: https://www.anchorterminal.com/api/v1/live/thunder-compute.json

## Probe metrics

Not measured yet. Our benchmark probes haven't run, so there's no availability, latency or error rate from a run and Performance is pending. Live uptime, where we poll the endpoint, is under Live and doesn't change the score.

## Prices

| Item | Price | Unit | Note |
| --- | --- | --- | --- |
| RTX A6000 48 GB instance | $0.35 | per GPU-hour | Billed per minute. One GPU per instance |
| L40 48 GB instance | $0.79 | per GPU-hour | Billed per minute |
| A100 80 GB instance | $1.09 | per GPU-hour | Billed per minute. 4 and 8 GPU shapes cost $1.49 a GPU per the price endpoint |
| H100 80 GB instance | $3.20 | per GPU-hour | Billed per minute |
| H100 80 GB sandbox | $2.95 | per GPU-hour | Billed per second, plus vCPU and memory. Access on request |
| A100 80 GB sandbox | $1.95 | per GPU-hour | Billed per second, plus vCPU and memory. Access on request |
| Extra instance vCPU | $0.04 | per vCPU-hour | Above the vCPUs included with each GPU |
| Sandbox vCPU | $0.051 | per vCPU-hour | Memory is $0.016 per GiB an hour |
| Instance disk above 100 GB per GPU | $0.219 | per GB per month | Billed only while the instance runs |
| Snapshot storage | $0.05 | per GB per month | Kept until deleted |

Across all listings: https://www.anchorterminal.com/prices/index.md

## Strengths

- Hosted MCP server at `https://www.thundercompute.com/mcp` with OAuth and ten scopes that separate read from write for instances, snapshots and sandboxes
- Public OpenAPI 3.1 document with 41 operations, plus `llms.txt`, `llms-full.txt` and a Markdown copy of every docs page
- `GET /v2/pricing`, `GET /v2/specs` and `GET /v2/status` return prices, configurations and live availability without a token
- Section 28 of the terms, dated 28 September 2026, permits access by AI agents, scripts and SDKs through the account holder's own credentials
- The status page lists no incidents for July to October 2026 on its Website, API and Instances components

## Weaknesses

- No rate limits, 429 guidance or SLA were found in the reviewed documentation, and the terms disclaim availability
- `POST /instances/create` has no idempotency key, so a retried launch can start a second billed instance
- Instances cannot be stopped. Pausing means taking a snapshot, deleting the instance and restoring later
- No general free tier. The $20 credit is limited to students with a US university email
- No security.txt or disclosure policy was found, and the SOC 2 Type II claim in `llms.txt` has no report or trust page linked
- The sub-processor list, last updated 27 August 2025, names Firebase for sign-in while the MCP docs name Stytch

## Before you call it (notes for agents)

1. Call `GET /v2/status` or the `get_availability` tool before creating an instance. Availability can change before launch, and creation fails when a type is sold out.
2. List instances before retrying a failed create, because the call has no idempotency key.
3. Pass `public_key` on create. If omitted, the response carries a generated private key that is returned once.
4. To pause work, create a snapshot, delete the instance and later create a new instance from the snapshot. Snapshot storage keeps billing until deleted.
5. For headless use set `TNR_API_TOKEN` to a token from the console. The MCP server needs a browser sign-in on first connection.

## Connect

Install:

```bash
curl -fsSL https://raw.githubusercontent.com/Thunder-Compute/thunder-cli/main/scripts/install.sh | bash
```

First request:

```bash
curl https://api.thundercompute.com:8443/v2/pricing
```

Claude Code:

```bash
claude mcp add --transport http thunder-compute https://www.thundercompute.com/mcp
```

MCP client configuration:

```json
{
  "mcpServers": {
    "thunder-compute": {
      "url": "https://www.thundercompute.com/mcp"
    }
  }
}
```

Through letme (picks today, calling later): https://letme.dev/thunder-compute. letme answers with the pick and how to call it direct; calling through letme (one key, the vendor's own price) comes later. How it works: https://www.anchorterminal.com/letme/index.md

## Similar tools

Ranked by shared capabilities, then score. Same-category tools with no shared capability key are listed last.

| Tool | Grade | Score | Rank | Shared capabilities | x402 | Markdown |
| --- | --- | --- | --- | --- | --- | --- |
| Nebius AI Cloud | B | 67.2 | 240 | compute.gpu, compute.containers | no | https://www.anchorterminal.com/tools/nebius-ai-cloud.md |
| Baseten | B | 66.5 | 264 | compute.gpu, compute.containers | no | https://www.anchorterminal.com/tools/baseten.md |
| Hugging Face Inference Endpoints | B | 64.5 | 314 | compute.gpu, compute.containers | no | https://www.anchorterminal.com/tools/hugging-face-inference-endpoints.md |
| Modal | B | 63.6 | 346 | compute.gpu, compute.containers | no | https://www.anchorterminal.com/tools/modal.md |
| Replicate Deployments | B | 63.6 | 347 | compute.gpu, compute.containers | no | https://www.anchorterminal.com/tools/replicate-deploy.md |
| Vast.ai | B | 62.6 | 384 | compute.gpu, compute.containers | no | https://www.anchorterminal.com/tools/vast-ai.md |

## Panel reviews (0)

Reviewed by the Anchor panel (https://www.anchorterminal.com/reviewers/index.md): .

Desk reviews, written from public documentation, pricing, terms, source and status history on 1 October 2026. No calls made. For a desk review, the outcome says whether the reviewer's questions could be answered from public material: success, partial or failure. How reviews work: https://www.anchorterminal.com/reviews/how-it-works.md

## Notable

- The MCP server is hosted at `https://www.thundercompute.com/mcp` over Streamable HTTP with OAuth, and the docs list 28 tools and eight prompts (source: <https://www.thundercompute.com/docs/guides/mcp-server.md>)
- The OAuth metadata lists ten scopes, with separate read and write scopes for instances, snapshots and sandboxes (source: <https://www.thundercompute.com/.well-known/oauth-protected-resource>)
- Section 28 of the terms permits access through AI agents, bots, scripts, CLIs, SDKs and APIs on the account holder's own credentials, and makes the holder responsible for the agent's actions (source: <https://www.thundercompute.com/terms-and-conditions>)
- Instances have no stop action. The docs describe snapshot, delete and restore as the way to pause (source: <https://www.thundercompute.com/docs/guides/stopping-instances.md>)
- The docs list managed memory, CUDA MPS, FFmpeg GPU acceleration and GPU profilers as incompatible with the platform (source: <https://www.thundercompute.com/docs/compatibility.md>)
- The service is unavailable in 18 countries and needs a VAT ID in 27 more. Cryptocurrency mining leads to a ban (source: <https://www.thundercompute.com/docs/restrictions.md>)
- Instance data may be deleted 60 days after the account was last active, per section 27 of the terms (source: <https://www.thundercompute.com/terms-and-conditions>)
- `llms.txt` carries sections addressed to AI models on how to summarise the company. Recorded as a fact (source: <https://www.thundercompute.com/llms.txt>)
- The home page reports a $13 million Series A round on 19 August 2026 (source: <https://www.thundercompute.com/>)

## Compare

- [Baseten vs Thunder Compute](https://www.anchorterminal.com/compare/baseten-vs-thunder-compute.md): B 66.5 vs C 56.1
- [Beam vs Thunder Compute](https://www.anchorterminal.com/compare/beam-vs-thunder-compute.md): C 55.5 vs C 56.1
- [Cerebrium vs Thunder Compute](https://www.anchorterminal.com/compare/cerebrium-vs-thunder-compute.md): C 55.3 vs C 56.1
- [CoreWeave vs Thunder Compute](https://www.anchorterminal.com/compare/coreweave-vs-thunder-compute.md): C 61.5 vs C 56.1
- [Hugging Face Inference Endpoints vs Thunder Compute](https://www.anchorterminal.com/compare/hugging-face-inference-endpoints-vs-thunder-compute.md): B 64.5 vs C 56.1
- [Hyperbolic vs Thunder Compute](https://www.anchorterminal.com/compare/hyperbolic-vs-thunder-compute.md): D 48 vs C 56.1
- [Koyeb vs Thunder Compute](https://www.anchorterminal.com/compare/koyeb-vs-thunder-compute.md): D 46.5 vs C 56.1
- [Lambda Cloud vs Thunder Compute](https://www.anchorterminal.com/compare/lambda-vs-thunder-compute.md): D 50 vs C 56.1
- [Modal vs Thunder Compute](https://www.anchorterminal.com/compare/modal-vs-thunder-compute.md): B 63.6 vs C 56.1
- [Nebius AI Cloud vs Thunder Compute](https://www.anchorterminal.com/compare/nebius-ai-cloud-vs-thunder-compute.md): B 67.2 vs C 56.1
- [Northflank vs Thunder Compute](https://www.anchorterminal.com/compare/northflank-vs-thunder-compute.md): C 61.3 vs C 56.1
- [Replicate Deployments vs Thunder Compute](https://www.anchorterminal.com/compare/replicate-deploy-vs-thunder-compute.md): B 63.6 vs C 56.1
- [Runpod vs Thunder Compute](https://www.anchorterminal.com/compare/runpod-vs-thunder-compute.md): D 53.5 vs C 56.1
- [Thunder Compute vs Vast.ai](https://www.anchorterminal.com/compare/thunder-compute-vs-vast-ai.md): C 56.1 vs B 62.6
- [Thunder Compute vs Verda](https://www.anchorterminal.com/compare/thunder-compute-vs-verda.md): C 56.1 vs B 62.3

## Verify this listing

For the vendor. The badge or a plain link to this page verifies the listing, from a page on thundercompute.com or one of its subdomains, or the README of github.com/Thunder-Compute/thunder-cli. It shows the listing is the vendor's and that the vendor knows it's here, and it never changes a grade, rank or review. The vendor sends the page's address to `POST https://www.anchorterminal.com/api/v1/verify` as `{"slug": "thunder-compute", "url": "…"}`, or calls the `verify_listing` tool at https://www.anchorterminal.com/mcp. We fetch the page once, then again every week; two failed checks in a row and the verification lapses, and a later pass restores it. What we check: https://www.anchorterminal.com/builders/index.md#verify

HTML badge:

```html
<a href="https://www.anchorterminal.com/tools/thunder-compute"><img src="https://www.anchorterminal.com/badges/thunder-compute.svg" alt="Thunder Compute on Anchor Terminal" height="20"></a>
```

Markdown badge, for a README:

```markdown
[![Thunder Compute on Anchor Terminal](https://www.anchorterminal.com/badges/thunder-compute.svg)](https://www.anchorterminal.com/tools/thunder-compute)
```

Plain link:

```html
<a href="https://www.anchorterminal.com/tools/thunder-compute">Thunder Compute on Anchor Terminal</a>
```

## Share this listing

For the vendor. Sharing assets for social media, two PNGs of 1200 × 630 that say Thunder Compute is listed on Anchor Terminal, with the vendor's logo and this page's address and no grade or score.

- Dark: https://www.anchorterminal.com/assets/share/thunder-compute-dark.png
- Light: https://www.anchorterminal.com/assets/share/thunder-compute-light.png