{"service":"HALOWERK modellwerk","endpoint":"/v1/model-router","method":"POST","content_type":"application/json","inputSchema":{"type":"object","required":["task"],"additionalProperties":false,"properties":{"task":{"type":"string","enum":["classification","extraction","summarization","translation","chat","coding","agent","reasoning","research","vision","long_context"],"description":"What the model has to do."},"budget_usd_per_call":{"type":"number","minimum":0,"description":"Ceiling per call. Models above it are excluded."},"speed_priority":{"type":"string","enum":["low","normal","high"],"default":"normal","description":"How much latency matters."},"input_tokens":{"type":"integer","minimum":0,"default":2000,"description":"Expected input size in tokens."},"input_text":{"type":"string","maxLength":100000,"description":"Sample input. Used to estimate input_tokens if that is not given."},"output_tokens":{"type":"integer","minimum":0,"description":"Expected output length. Defaults to a value typical for the task."},"allowed_vendors":{"type":"array","items":{"type":"string"},"description":"Restrict to these vendors, e.g. for a data-residency or procurement rule."},"excluded_vendors":{"type":"array","items":{"type":"string"},"description":"Exclude these vendors."},"require_tags":{"type":"array","items":{"type":"string"},"description":"Capabilities the model must carry, e.g. vision."},"weights":{"type":"object","additionalProperties":false,"description":"Weighting of cost, capability and speed. Normalised to sum 1.","properties":{"cost":{"type":"number","minimum":0,"default":0.4},"capability":{"type":"number","minimum":0,"default":0.4},"speed":{"type":"number","minimum":0,"default":0.2}}}}},"requestExample":{"task":"classification","speed_priority":"normal","input_tokens":2000},"operations":[{"method":"POST","inputSchema":{"type":"object","required":["task"],"additionalProperties":false,"properties":{"task":{"type":"string","enum":["classification","extraction","summarization","translation","chat","coding","agent","reasoning","research","vision","long_context"],"description":"What the model has to do."},"budget_usd_per_call":{"type":"number","minimum":0,"description":"Ceiling per call. Models above it are excluded."},"speed_priority":{"type":"string","enum":["low","normal","high"],"default":"normal","description":"How much latency matters."},"input_tokens":{"type":"integer","minimum":0,"default":2000,"description":"Expected input size in tokens."},"input_text":{"type":"string","maxLength":100000,"description":"Sample input. Used to estimate input_tokens if that is not given."},"output_tokens":{"type":"integer","minimum":0,"description":"Expected output length. Defaults to a value typical for the task."},"allowed_vendors":{"type":"array","items":{"type":"string"},"description":"Restrict to these vendors, e.g. for a data-residency or procurement rule."},"excluded_vendors":{"type":"array","items":{"type":"string"},"description":"Exclude these vendors."},"require_tags":{"type":"array","items":{"type":"string"},"description":"Capabilities the model must carry, e.g. vision."},"weights":{"type":"object","additionalProperties":false,"description":"Weighting of cost, capability and speed. Normalised to sum 1.","properties":{"cost":{"type":"number","minimum":0,"default":0.4},"capability":{"type":"number","minimum":0,"default":0.4},"speed":{"type":"number","minimum":0,"default":0.2}}}}},"requestExample":{"task":"classification","speed_priority":"normal","input_tokens":2000}}],"example_request":{"task":"classification","speed_priority":"normal","input_tokens":2000},"serviceName":"model_router","tags":["llm","routing","selection"],"outputSchema":{"type":"object"},"outputExample":{"task":"classification","task_profile":{"requires":["speed","cheap"],"reasoning_load":1,"typical_output_tokens":20},"recommendation":{"model":"gpt-5-mini","vendor":"openai","cost_usd_per_call":0.00054,"why":"Best weighted result among 10 eligible models: Kosten 100 von 100, Eignung 100 von 100 (speed, cheap), Tempo 100 von 100."},"ranked":[{"model":"gpt-5-mini","vendor":"openai","score":1,"subscores":{"cost":1,"capability":1,"speed_proxy":1},"cost_usd_per_call":0.00054,"matched_strengths":["speed","cheap"],"missing_strengths":[],"context_window":400000,"context_headroom":397980,"price_table_date":"2026-05-01","rank":1},{"model":"gemini-2.5-flash","vendor":"google","score":0.9958,"subscores":{"cost":0.9946,"capability":1,"speed_proxy":0.9896},"cost_usd_per_call":0.00065,"matched_strengths":["speed","cheap"],"missing_strengths":[],"context_window":1048576,"context_headroom":1046556,"price_table_date":"2026-05-01","rank":2},{"model":"claude-haiku-4-5","vendor":"anthropic","score":0.957,"subscores":{"cost":0.9238,"capability":1,"speed_proxy":0.9375},"cost_usd_per_call":0.0021,"matched_strengths":["speed","cheap"],"missing_strengths":[],"context_window":200000,"context_headroom":197980,"price_table_date":"2026-06-24","rank":3},{"model":"gpt-5","vendor":"openai","score":0.5244,"subscores":{"cost":0.8944,"capability":0,"speed_proxy":0.8333},"cost_usd_per_call":0.0027,"matched_strengths":[],"missing_strengths":["speed","cheap"],"context_window":400000,"context_headroom":397980,"price_table_date":"2026-05-01","rank":4},{"model":"gemini-2.5-pro","vendor":"google","score":0.5244,"subscores":{"cost":0.8944,"capability":0,"speed_proxy":0.8333},"cost_usd_per_call":0.0027,"matched_strengths":[],"missing_strengths":["speed","cheap"],"context_window":1048576,"context_headroom":1046556,"price_table_date":"2026-05-01","rank":5},{"model":"claude-sonnet-5","vendor":"anthropic","score":0.4332,"subscores":{"cost":0.7185,"capability":0,"speed_proxy":0.7292},"cost_usd_per_call":0.0063,"matched_strengths":[],"missing_strengths":["speed","cheap"],"context_window":1000000,"context_headroom":997980,"price_table_date":"2026-06-24","rank":6},{"model":"claude-sonnet-4-6","vendor":"anthropic","score":0.4332,"subscores":{"cost":0.7185,"capability":0,"speed_proxy":0.7292},"cost_usd_per_call":0.0063,"matched_strengths":[],"missing_strengths":["speed","cheap"],"context_window":1000000,"context_headroom":997980,"price_table_date":"2026-06-24","rank":7},{"model":"claude-opus-5","vendor":"anthropic","score":0.3094,"subscores":{"cost":0.5132,"capability":0,"speed_proxy":0.5208},"cost_usd_per_call":0.0105,"matched_strengths":[],"missing_strengths":["speed","cheap"],"context_window":1000000,"context_headroom":997980,"price_table_date":"2026-06-24","rank":8},{"model":"claude-opus-4-8","vendor":"anthropic","score":0.3094,"subscores":{"cost":0.5132,"capability":0,"speed_proxy":0.5208},"cost_usd_per_call":0.0105,"matched_strengths":[],"missing_strengths":["speed","cheap"],"context_window":1000000,"context_headroom":997980,"price_table_date":"2026-06-24","rank":9},{"model":"claude-fable-5","vendor":"anthropic","score":0,"subscores":{"cost":0,"capability":0,"speed_proxy":0},"cost_usd_per_call":0.021,"matched_strengths":[],"missing_strengths":["speed","cheap"],"context_window":1000000,"context_headroom":997980,"price_table_date":"2026-06-24","rank":10}],"runner_up":{"model":"gemini-2.5-flash","score":0.9958,"cost_difference_usd":0.00011},"excluded":[],"tokens":{"input":2000,"output":20,"total":2020,"estimated":false},"weights_used":{"cost":0.4,"capability":0.4,"speed":0.2,"speed_priority":"normal"},"limits":"Heuristik ueber veroeffentlichte Preise und Faehigkeitsmerkmale, kein Leistungsvergleich. speed_proxy ist vom Ausgabepreis abgeleitet, nicht gemessen. Fuer qualitaetskritische Faelle ist das eine Vorauswahl zum Testen, kein Urteil."}}