Dellbot LLM API
activex402-gated LLM / TTS / STT / image / embedding inference powered by Venice AI. Pay per request in USDC on Base mainnet. Thirteen tiers: chat-nano/chat/chat-pro/chat-max, tts-nano/tts-pro, image-nano/image/image-pro, embeddings, upscale, background-remove, transcribe. Indexed via CDP x402 Bazaar.
Settled via Coinbase.
- Transactions · 30d
- 4
- Volume · 30d
- $0.012
- Unique buyers · 30d
- 3
- Uptime · 30d
- 94.8%
- Latency p50
- 367ms
- Reported calls · 30d
- 4
Endpoints (13 live)
POST/chat— Mid-tier OpenAI-compatible chat completion. Qwen3 235B (instruct/thinking/VL/next), DeepSeek V3.2, GLM 5.2, venice-uncensored 1.2, GLM 4.6/4.7, Kimi K2.5, MiniMax M3 Preview, Nemotron 3 Ultra 550B, DeepSeek V4.1 Flash, Gemma 4 26B. Supports tools/tool_choice and stream:true SSE. (0.003 USDC on Base)POST/tts-nano— Cheapest TTS on x402 Bazaar. POST text, get back MP3 audio. tts-kokoro (open-weights, fast, multi-voice). 1000-char input cap. Returns audio/mpeg binary directly — no JSON wrapper to decode. (0.005 USDC on Base)POST/background-remove— AI background removal. POST {image (base64)} or {image_url}. Returns a binary image/png with transparent background. Source < 25MB. (0.06 USDC on Base)POST/transcribe— Speech-to-text on x402 Bazaar. POST multipart/form-data with a `file` audio upload (WAV/FLAC/M4A/MP3/OGG/WEBM). Whisper Large v3 (default), NVIDIA Parakeet, fal Wizper, xAI STT. Optional model/response_format/timestamps/language fields. Returns JSON {text, duration, timestamps?}. Hard cap 300s of audio per request (longer → 400, so check duration before paying). (0.05 USDC on Base)POST/upscale— AI image upscale & enhance. POST {image (base64), scale 1-4, enhance?, enhanceCreativity?, enhancePrompt?, replication?}. Scale 1 requires enhance=true (enhancer-only). Returns binary image/png. Source image must be >= 65536px and < 25MB. (0.15 USDC on Base)POST/embeddings— Text embeddings on x402 Bazaar. OpenAI-compatible /embeddings. POST {input, model?} where input is a string or array of strings (combined length capped at 24000 chars). Default BGE-M3; also Qwen3 8B/0.6B, multilingual E5, Nemotron VL, OpenAI text-embedding-3 small/large, BGE-en-icl. Returns JSON {data:[{embedding:[...]}], usage}. (0.005 USDC on Base)POST/chat-max— Frontier-flagship reasoning. Claude Opus 4.5 / 4.6 / 4.7 / 4.8 and OpenAI gpt-55. Output is hard-capped at 4096 tokens to keep the fixed price margin-safe on $30-$37.50/1M models. OpenAI-compatible. Supports tools/tool_choice and stream:true SSE. (0.2 USDC on Base)POST/image-pro— Premium image generation. Flux 2 Pro (default) / Flux 2 Max, Hunyuan v3, Qwen Image 2 / 2-Pro, Recraft v4, Seedream v4 / v5-Lite, Wan 2.7 / Wan 2.7 Pro, ImagineArt 1.5 Pro. Returns JSON {images: [base64]}. (0.15 USDC on Base)POST/tts-pro— Premium TTS. ElevenLabs Turbo v2.5 (default), Orpheus, Chatterbox HD, Grok xAI voice, Inworld 1.5 Max. 500-char input cap. Returns audio/mpeg binary. (0.08 USDC on Base)POST/image— Mid-tier image generation. Flux 2 Pro (default) and Wan 2.7 at a third of the pro price. Returns JSON {images: [base64]}. (0.05 USDC on Base)POST/chat-pro— Pro-tier reasoning & code. Claude Sonnet 5, Kimi K2.7 Code, Qwen3.7 Plus, Qwen3-Coder 480B, Grok-4 family (4-20, 4-3), DeepSeek V4, OpenAI gpt-54/gpt-52/gpt-53-codex, GLM 5.1. OpenAI-compatible. Supports tools/tool_choice and stream:true SSE. (0.01 USDC on Base)POST/chat-nano— Cheapest LLM on CDP x402 Bazaar. OpenAI-compatible chat completion. Default OpenAI GPT-OSS 120B; also offers Llama 3.2 3B, Mistral Small 24B, Gemma 3 27B, Qwen3.5 9B, and Nemotron Nano 30B (all via Venice AI). 512-token output cap, 6000-char input cap. Supports tools/tool_choice and stream:true SSE. (0.001 USDC on Base)POST/image-nano— Cheapest image generation on x402 Bazaar. Stable Diffusion 3.5 (default), Lustify SDXL/v7/v8, Qwen Image, Chroma, WAI Illustrious, Z-Image Turbo. Returns JSON {images: [base64-webp]}. (0.02 USDC on Base)
First seen · last seen · last active