Synap by Linkrra
activeOpenAI-compatible inference for agents. Pay per call in USDC on Base—no account or API key. Dedicated GPUs, streaming responses, and model-level pricing.
Settled via Coinbase.
- Transactions · 30d
- 335
- Volume · 30d
- $0.3477
- Unique buyers · 30d
- 25
- Uptime · 30d
- 100.0%
- Latency p50
- 150ms
- Reported calls · 30d
- 228
Endpoints (35 live)
POST/api/billing/x402/models/gpt-oss-20b/chat/completions— Synap by Linkrra · OpenAI gpt-oss-20b — fast + cheap general model. OpenAI-compatible inference: 128K context, dedicated GPUs we operate, native 4-bit MXFP4. Pay per call in USDC on any network offered in the payment challenge; no account, no API key. First call free per new wallet (cheap models, up to 512 tokens, non-streaming, daily cap). From $0.001 per call. Streaming supported. (0.001 USDC on Base)POST/api/billing/x402/tts— Linkrra Voice — text-to-speech with WAV output and automatic Hebrew nikud. Pay per call in USDC on any network offered in the payment challenge; no account or API key. From $0.01 per call. (0.01 USDC on Base)POST/api/billing/x402/models/qwen3-vl-30b-a3b-instruct/chat/completions— Synap by Linkrra · Qwen3 VL 30B A3B Instruct — vision. OpenAI-compatible inference: 64K context, dedicated GPUs we operate, 4-bit AWQ. Pay per call in USDC on any network offered in the payment challenge; no account, no API key. From $0.001 per call. Streaming supported. (0.001 USDC on Solana)POST/api/billing/x402/models/qwen3-coder-30b/chat/completions— Synap by Linkrra · Qwen3 Coder 30B A3B Instruct — coding, everyday. OpenAI-compatible inference: 96K context, dedicated GPUs we operate, 4-bit AWQ. Pay per call in USDC on any network offered in the payment challenge; no account, no API key. From $0.001 per call. Streaming supported. (0.001 USDC on Base)POST/api/billing/x402/models/qwen3-30b-a3b/chat/completions— Synap by Linkrra · Qwen3 30B A3B Instruct 2507 — general, high-throughput. OpenAI-compatible inference: 96K context, dedicated GPUs we operate, 4-bit AWQ. Pay per call in USDC on any network offered in the payment challenge; no account, no API key. First call free per new wallet (cheap models, up to 512 tokens, non-streaming, daily cap). From $0.001 per call. Streaming supported. (0.001 USDC on Base)POST/api/billing/x402/models/muse-glimmer-30b/chat/completions— Synap by Linkrra · Meta Muse Glimmer 30B — agentic reasoning + tool use (text). OpenAI-compatible inference: 128K context, dedicated GPUs we operate, 4-bit AWQ. Pay per call in USDC on any network offered in the payment challenge; no account, no API key. From $0.001 per call. Streaming supported. (0.001 USDC on Base)POST/api/billing/x402/models/gemma-4-26b-a4b-it/chat/completions— Synap by Linkrra · Google Gemma 4 26B A4B Instruct — fast general model, long context. OpenAI-compatible inference: 128K context, dedicated GPUs we operate, 4-bit AWQ. Pay per call in USDC on any network offered in the payment challenge; no account, no API key. From $0.001 per call. Streaming supported. (0.001 USDC on Base)POST/api/billing/x402/chat/completions— Synap by Linkrra — OpenAI-compatible chat completions for coding and agent workloads on dedicated GPUs. Pay per call in USDC on any network offered in the payment challenge; no account, no API key. First call free per new wallet (cheap models, up to 512 tokens, non-streaming, daily cap). Self-hosted open models use 4-bit weights. From $0.001 per call. Streaming supported: https://synap402.com (0.001 USDC on Base)POST/api/billing/x402/topup— Synap credits — fund an inference wallet with USDC on any network offered in the payment challenge. $1–$1000 per top-up; $0.2273/M credits. (1 USDC on Base)POST/api/billing/x402/tasks/review-code— Synap402 - code review: send a unified diff or source code, get a summary and a JSON list of issues (severity, file, line, message, suggestion) - bugs, security, edge cases, no style nits. Invalid model output = no charge. Open models on dedicated GPUs. Pay per call in USDC on any network in the payment challenge; no account, no API key. (0.001 USDC on Solana)GET/api/billing/x402/models/qwen3-embedding-0.6b/chat/completions(0.001 USDC on Solana)POST/api/billing/x402/tasks/code-patch— Synap by Linkrra - code patch: send a bounded change request and the source files it may edit; get a server-validated JSON result containing a unified diff, explanation, and tests. The patch may name only supplied files. Invalid model output = no charge. Pay per call in USDC via x402; no account or API key. (0.00109 USDC on Base)POST/api/billing/x402/tasks/structured-extraction— Synap by Linkrra - structured extraction: send source text and a strict JSON Schema; get data validated against that schema server-side. The request contract rejects unknown fields, and invalid model output is repaired once or returned uncharged. Pay per call in USDC via x402; no account or API key. (0.001 USDC on Base)POST/api/billing/x402/tasks/vision-inspection— Synap by Linkrra - vision inspection: send a public image URL or base64 image and receive server-validated JSON with summary, objects, visible text, and a quality/category assessment. Unknown input fields and malformed model output are rejected; task failure is not charged. Pay per call in USDC via x402. (0.020126 USDC on Base)POST/api/billing/x402/tasks/describe-image— Synap by Linkrra - describe an image: send image_url (fetched server-side, public URLs only) or image_base64 (PNG, JPEG, GIF, WebP, 4 MB) and an optional question; get a precise description with visible text transcribed. Qwen3-VL on dedicated GPUs. Fetch fails = no charge. Pay per call in USDC; no account, no API key. (0.009471 USDC on Solana)POST/api/billing/x402/tasks/extract-json— Synap by Linkrra - JSON extraction: send text and a JSON Schema, get data validated against that schema server-side (one automatic repair retry). Output still invalid = 422 and nothing is charged. Open models on dedicated GPUs. Pay per call in USDC on any network in the payment challenge; no account, no API key. (0.001 USDC on Base)POST/api/billing/x402/tasks/summarize-url— Synap by Linkrra - summarize a web page: send a URL, we fetch it server-side (public http/https, HTML or plain text, 2 MB, 10 s) and return a faithful summary plus the page title. Fetch fails = no charge. Open models on dedicated GPUs. Pay per call in USDC on any network in the payment challenge; no account, no API key. (0.002641 USDC on Base)GET/api/billing/x402/tasks/vision-inspection(0.020479 USDC on Base)GET/api/billing/x402/models/qwen3.5-122b-a10b/chat/completions(0.001 USDC on Base)GET/api/billing/x402/tasks/structured-extraction(0.02317 USDC on Base)
+15 more endpoints.
First seen · last seen · last active