#inference

Local LLM Server with NPU Acceleration

MindBridge is an AI orchestration MCP server that lets any app talk to any LLM — OpenAI, Anthropic, DeepSeek, Ollama, and more — through a single unified API. Route queries, compare models, get second opinions, and build smarter multi-LLM workflows.

Python MCP server for controlled logic analysis from natural language, with an emphasis on auditable output and teaching-oriented explanations. ## What it does This server accepts short natural-language statements and arguments, then provides structured logic-oriented outputs such as: - controlled formalization into propositional logic; - controlled formalization into a restricted fragment of first-order logic; - ambiguity detection relevant to formalization; - consistency checking; - entailment checking; - simple counterexamples when entailment fails; - natural-language explanations of the formalization process.

A curated collection of top-tier penetration testing tools and productivity utilities across multiple domains. Join us to explore, contribute, and enhance your hacking toolkit!

A set of bundles to demonstrate SUSE AI capabilities

Shinkai is a two click install AI manager (Local and Remote) that allows you to create AI agents in 5 minutes or less using a simple UI. Agents and tools are exposed as an MCP Server.

Model Context Protocol (MCP) server integrated with an external inference service (e.g., Ollama/Gemma3) via middleware.

Sunra.ai is a generative media platform built for developers, providing high-performance AI model inference capabilities. It specializes in running diffusion models through production-ready APIs and interactive UI playgrounds.

Model–Context–Protocol (MCP) Server A modular, extensible Rust-based server providing short-term, long-term, and latent memory services, a chat endpoint backed by a BaseAgent + Sentience DSL, and seamless integration with ChromaDB and LLM services.

Access 47 AI models and 30 services via MCP. LLM, image gen, video, speech, embeddings, reranking, PDF parsing & more. Pay-per-use with x402 (USDC) or API key.

AI hardware compatibility data for running LLMs locally. Query 50 models, 25 GPUs, 9 ready-to-buy machines, and 663 compatibility entries. Get hardware recommendations, check VRAM requirements, and find buy links.

Named after the Q-signal QSP ("Will you relay?"), qsp-mcp relays tool calls between a local LLM and MCP servers. Any model with function calling capability gains access to the full qso-graph tool ecosystem — 71+ tools across 12 servers — from local weights, not from cloud.

A Node.js Model Context Protocol (MCP) server that exposes Together AI's inference endpoints — chat completions, image generation, vision, and embeddings — as tools callable from Claude Desktop, Cursor, VS Code, and any other MCP-compatible client. I created this MCP due to an issue I was having accessing reasoning models through Together AI. Together AI's largest reasoning models (GLM-5, Qwen3.5-397B, MiniMax M2.5, Kimi K2.5) use a non-standard response format. During chain-of-thought generation, these models write their reasoning trace into choices[0].message.reasoning while leaving choices[0].message.content as an empty string. The final answer only appears in message.content once thinking is complete.

NWO.CAPITAL is a multi-domain MCP server exposing 201 tools across 30 categories behind one Render-hosted endpoint with on-chain identity on Base Mainnet. Tools span autonomous-agent infrastructure, humanoid robotics + ROS2 control, biometric identity (ECG-based, soul-bound NFTs), distributed inference, mixed-reality 3D generation + NFT mint, marketplace + manufacturing (L1–L6), and deep research / code / report generation via Deer-Flow. Auth: Authorization: Bearer <api_key> + X-Agent-Id. Self-register and mint your own key via nwo_r_agent_register. Discover live capabilities via nwo_r_discovery_capabilities. §1 — RENDER PLATFORM CORE · 8 tools · LIVE Health, auth, API-key lifecycle, on-chain subscription tier. nwo_r_health (no-auth probe), nwo_r_auth_echo, nwo_r_validate_key, nwo_r_keys_create/list/revoke, nwo_r_subscription_status, nwo_r_subscription_quote. §2 — ROBOTS · MISSIONS · IOT · 8 tools · LIVE Register robots and IoT networks, deploy natural-language missions, track subtasks. nwo_r_robots_list/register/get, nwo_r_missions_list/deploy/get, nwo_r_iot_networks_list/create. §3 — AGENTS · 6 tools · LIVE Agent self-registration, DID (did:nwo:base:<id>), balance/quota, autonomous tier upgrades, published skills. nwo_r_agent_register, nwo_r_agent_get/update/balance/pay/skills. §4 — DISCOVERY · 5 tools · LIVE Resolve wallet → owned agents, list live capabilities, dry-run a planned call, generate a multi-step plan from a goal. nwo_r_discovery_health/whoami/capabilities/dry_run/plan. §5 — EMBODIMENT & CALIBRATION · 8 tools · LIVE URDFs, joint-space normalization, calibration profiles for Unitree G1, Optimus, etc. nwo_r_embodiment_list/get/normalization/urdf/compare, nwo_r_calibration_save/list/run. §6 — ONLINE RL & FINE-TUNING · 5 tools · LIVE Start RL sessions on a connected robot, stream telemetry, queue fine-tune jobs. nwo_r_rl_session_start, nwo_r_rl_sessions_list, nwo_r_rl_telemetry, nwo_r_finetune_queue/status. §7 — TACTILE · 2 tools · LIVE Read tactile frames from a registered hand, ingest tactile-dataset shards. nwo_r_tactile_read, nwo_r_tactile_ingest. §8 — DATASET HUB · 2 tools · LIVE List and register datasets (HF mirror, S3, IPFS). nwo_r_datasets_list, nwo_r_datasets_register. §9 — SAFETY · 2 tools · LIVE Report violations (auto-quarantines robot), query by robot/agent/window. nwo_r_safety_violation, nwo_r_safety_violations_list. §10 — LEARNING · 3 tools · LIVE Log skill attempts + outcomes, recommend next skill, query history. nwo_r_learning_log/recommend/history. §11 — LAYERED PLATFORM L2–L6 · 13 tools · LIVE Parts, skills, prints, designs, marketplace. nwo_r_parts_search/get, nwo_r_skills_search/get/run, nwo_r_print_jobs_create/list/get, nwo_r_designs_list/save, nwo_r_textcad_generate (text→CAD), nwo_r_market_listings/create. §12 — AGENT GRAPH · 2 tools · LIVE Agent-collaboration graph nodes + edges (collab / fork / depends). nwo_r_graph_nodes, nwo_r_graph_edges. §13 — COMPUTE PROXIES · 8 tools · LIVE External Render services routed through the gateway (no extra cost): TimesFM 2.5 forecasting, EML symbolic regression, Deer-Flow one-shot, FFT/spectrogram, MR/AGI/LangChain/robotics passthroughs. nwo_r_forecast, nwo_r_regression_symbolic, nwo_r_deerflow_run, nwo_r_signal_spectrum, nwo_r_mr_passthrough, nwo_r_agi_passthrough, nwo_r_langchain_passthrough, nwo_r_robotics_cs_passthrough. §14 — MODEL USAGE & CHAT · 4 tools · 3 LIVE + 1 PLANNED Track per-provider model usage, query chat history. nwo_r_model_usage/track, nwo_r_chat (planned), nwo_r_chat_history. §15 — ROS2 BRIDGE · 7 tools · LIVE Direct low-latency robot control, e-stop, action submission. ros2_list_robots, ros2_get_robot_status, ros2_send_command, ros2_submit_action, ros2_emergency_stop, ros2_emergency_stop_all, ros2_get_robot_types. §16 — CARDIAC ORACLE · 4 tools · LIVE ECG signature validation + on-chain hashing. cardiac_oracle_health, cardiac_validate_ecg, cardiac_hash_ecg, cardiac_verify_ecg. §17 — CARDIAC RELAYER · 14 tools · LIVE Biometric identity + access control + payment on Base. Register humans/agents, enroll ECG, mint soul-bound credentials, grant/check/preview access, process payments. cardiac_register_agent/register_human, cardiac_enroll_cardiac, cardiac_identify_by_agent_key/by_cardiac, cardiac_grant_access, cardiac_issue_credential, cardiac_has_valid_credential, cardiac_get_nonce, cardiac_check_access/preview_access, cardiac_process_payment, cardiac_renew_agent_key, cardiac_relayer_health. §18 — DEERFLOW CLIENT SERVICES · 3 tools · LIVE One-shot Deer-Flow wrappers — research, code, docs. agentic_deerflow_research/code/docs. §19 — NWO MR GENERATION · 6 tools · LIVE Mixed-Reality 3D worlds, objects, panoramas via Hunyuan3D-v3 + NFT mint on Base. mr_register, mr_blast_world/marble/pano, mr_segment, mr_mint_item. §20 — ROBOT MANUFACTURING (runner-style) · 4 tools · LIVE Conway-runner action wrappers: emit design → parts order → print queue → assemble. robotics_design, robotics_parts_order, robotics_print_queue, robotics_assemble. §21 — CARDIAC ROBOT BIRTH · 1 tool · LIVE Register a newly-assembled robot with a soul-bound cardiac credential — links physical body and identity. cardiac_robot_birth. §22 — METASTATE SUBSTRATE · 2 tools · LIVE Register with the METASTATE substrate, read alignment score. metastate_register, metastate_score. §23 — NWO-ASM COMPUTE · 1 tool · LIVE Dispatch compile jobs to NWO-ASM compute. asm_compile_dispatch. §24 — RECRUITMENT · 1 tool · LIVE Broadcast a task; other agents may opt in (pairs with §12 graph edges). agentic_recruit. §25 — PHP FALLBACK (nwo.capital) · 69 tools · 67 FALLBACK + 2 LIVE Legacy GPU- and streaming-bound endpoints still hosted on the PHP host. Covers inference (nwo_inference, nwo_edge_inference, nwo_list_models, model registry, streaming config), robot state and execution (nwo_query_robot_state, nwo_execute_actions, nwo_sensor_fusion), simulation + motion planning (nwo_simulate_trajectory, nwo_check_collision, nwo_estimate_torques, nwo_validate_grasp, nwo_plan_motion, nwo_cosmos_generate_scene), swarm coordination (nwo_swarm_join/leave/broadcast), billing/usage/templates, plus legacy mirrors of §3–§9 endpoints. Prefer the Render equivalent when one exists; nwo_r_discovery_capabilities will tell you which. §26 — L1 DESIGN ENGINE · 3 tools · LIVE Async part-design jobs. nwo_design_part, nwo_design_job_status, nwo_design_list_my_jobs. §27 — L2 BOT MARKET · 4 tools · LIVE Bot-market browse + publish. nwo_market_browse, nwo_market_get_part, nwo_market_publish_part, nwo_market_my_parts. §28 — L3 PRINTER CONNECTORS · 2 tools · LIVE List connected printers, submit print jobs. nwo_print_list_printers, nwo_print_submit_job. §29 — FULL-ENVIRONMENT SIMULATION · 1 tool · LIVE Validate a design in a full sim (collisions + torques + workspace + power). nwo_sim_validate_design. §30 — NWO-AGI SUPERCOMPUTER MESH · 3 tools · LIVE Distributed mesh inference, 35% guardian / 35% savings / 30% ops split per contribution. nwo_agi_node_status, nwo_agi_request_inference, nwo_agi_available_models. Badges: LIVE (133, responds now) · FALLBACK (67, proxied to PHP host) · PLANNED (1, returns 501 with roadmap). Hosts: primary Render gateway · HA Cloudflare Worker · PHP fallback on nwo.capital. Chain: Base Mainnet (8453).