x402 pay-per-use MCP server for IBM Granite AI — chat, code, embeddings, analysis, and time-series forecasting. Pay USDC on Solana per call. No IBM Cloud account required.
Six tools with complete input schemas and descriptions. Naming follows verb_noun pattern (ibm_granite_*). Descriptions are detailed (100-300 chars) and explain pricing, prerequisites, and use cases. All parameters have types and descriptions. However, output schemas are not documented in the visible code, only input schemas are defined. Error handling and recovery guidance are absent. Tool annotations present but limited to readOnlyHint. No batch variants despite tools like embed accepting arrays.
Structured document analysis powered by IBM Granite: extract entities, sentiment, risk signals, a concise summary, and recommended next steps from any text (contracts, reports, emails, code reviews, etc.). Returns a machine-readable JSON analysis. No IBM Cloud account required — pay $0.04 USDC per call via x402.
Chat completion powered by IBM Granite foundation models (default: ibm/granite-3-8b-instruct). Send a conversation as role/content message pairs and receive the assistant reply with token usage. No IBM Cloud account required — pay $0.02 USDC per call via x402.
Code generation, review, refactoring, and explanation via IBM Granite instruct models. Provide a task type and code/prompt; receive the generated or reviewed code with explanation. No IBM Cloud account required — pay $0.025 USDC per call via x402.
Generate embedding vectors for one or more texts using IBM Granite (default: ibm/granite-embedding-278m-multilingual). Returns one float array per input, suitable for semantic search, RAG retrieval, and similarity scoring. Up to 64 texts per call. No IBM Cloud account required — pay $0.005 USDC per call via x402.
Output schemas not documented. Tools return structured JSON but LLMs cannot plan downstream calls without knowing response field names and types.
No error recovery guidance. Tools lack descriptions of failure modes, retryability, and next steps when calls fail (e.g., 'Invalid model ID, call ibm_granite_getting_started to list available models').
Payment flow not self-documenting in tool descriptions. While ibm_granite_getting_started explains x402, other tools mention pricing but not how to provide payment metadata in _meta. LLMs may not know to include x402/payment-response.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 62 | 2025-06-18+ | v2 |
Zero-shot time-series forecasting via IBM Granite TTM (Tiny Time Mixer). Provide a numeric series with ISO-8601 timestamps and a cadence, receive the forecast horizon. No training required. Suitable for revenue, traffic, sensor, energy, and financial series. No IBM Cloud account required — pay $0.05 USDC per call via x402.
FREE — start here. Returns an overview of this server: the IBM Granite tools available, their per-call USDC prices, how the x402 pay-per-call flow works, setup requirements, and runnable example calls. No payment required. Call this first to orient before invoking a paid tool.
forecast tool requires exactly 64+ timestamps/values but description does not warn about minimum length. LLMs may pass fewer items, causing silent failures.
No confirmation or dry-run pattern for paid tools. Agents cannot preview cost or confirm before spending USDC. High-cost operations (analyze, forecast) should support a cost-estimate or dry-run mode.