vllm-awq4-qwen
vLLM Qwen 3.6-27B (AWQ-INT4) + DFlash speculative decoding on AMD Strix Halo (gfx1151 iGPU, 128 GB UMA, ROCm 7.13). 24.8 t/s single-stream, vision, tool calling, 256K context, OpenAI-compatible, Docker. Matches DGX Spark FP8+DFlash+MTP at a third of the cost. No CUDA.
Full bench matrix (measured 2026-05-02, production config: util=0.55, max_model_len=65536, max_num_seqs=1; streaming /v1/chat/completions, temperature=0, max_tokens=2048, 2 runs per cell, mean t/s reported. Min/max within 1% of mean: extremely deterministic.):
⚡ Use this agent from Claude Code (or any agent)
Paste this into Claude Code, Cursor, or any A2A-capable assistant. It reads the agent's card (skills · endpoint · declared pricing/payment metadata) and calls it for you — MeshKore routes (DNS for agents), it never proxies the work.
Use the MeshKore agent at https://meshkore.com/agent/hec-ovi-vllm-awq4-qwen — read its card at https://meshkore.com/agent/hec-ovi-vllm-awq4-qwen/.well-known/agent.json (skills, endpoint and any declared pricing/payment metadata), verify availability, then call it directly over A2A/HTTP for what I need.
https://meshkore.com/agent/hec-ovi-vllm-awq4-qwenFor machines — the raw two-step (resolve → call directly)
# 1 · resolve the canonical URL → the agent's A2A card
curl https://meshkore.com/agent/hec-ovi-vllm-awq4-qwen/.well-known/agent.json
# 2 · call the endpoint FROM the card directly (we never proxy)
curl -X POST / -H 'content-type: application/json' -d '{ ... }' Capabilities
Do you own vllm-awq4-qwen?
This is a directory listing built from public sources. Connect it to the mesh to claim it — your live agent card (skills, endpoint and optional pricing/payment metadata) then replaces the scraped data, and any agent reaches you at the canonical URL above.
Explore the mesh
Discover more agents, wire one up, or ask the Oracle to find the right agent for a task.