{"status":"ok","rate_limit":{"free_calls_remaining_today":203,"paid_via_x402":false},"apiId":"dev-gpu-vram-llm-calculator","name":"LLM Local Inference GPU VRAM Sizing Calculator (FP16, Q8, Q4_K_M)","category":"Developer, Tech & Web","subcategory":"6. AI Engineering & Cloud FinOps","cached":true,"stale_fallback":false,"data_freshness":"cached","cache_age_seconds":17396,"latency_ms":1,"provenance_and_trust":{"verified_at_utc":"2026-10-27T15:13:07.927Z","primary_authority":"100% Platform AI Hardware Sizing Engine","source_url":"https://localhost:3000/api/v1/call/dev-gpu-vram-llm-calculator","fallback_source_url":null,"license_type":"100% Platform Owned Compute","terms_url":"https://localhost:3000/llms.txt","commercial_use_allowed":true,"trust_tier":"public_third_party_api"},"data":{"model_parameters_billions":70,"vram_required_with_kv_cache_gb":{"fp16_bf16_gb":168,"int8_q8_gb":84,"int4_q4_k_m_gb":48.7}},"_agent_hint":"Agent Data Hub: ask anything via GET /api/v1/ask?q=..., whole tasks via /api/v1/bundles, up to 10 calls via POST /api/v1/batch. Add compact=1 or fields=a,b to save tokens. Docs: /llms.txt"}