omnius 1.0.847 → 1.0.849

package/docs/work-orders/evidence/keyless-provider-survey-20261012.json

@@ -0,0 +1,1327 @@
+{
+  "surveyDate": "2026-10-12",
+  "generatedAtUtc": "2026-10-13T02:45:33.082706+00:00",
+  "scope": "Public primary documentation, public metadata, bounded synthetic remote probes only. No local inference, services, private conversation, real Telegram, account/user keys or challenge bypass.",
+  "registry": {
+    "url": "https://robit-man.github.io/inference/",
+    "reviewedAt": "2026-09-04",
+    "providerCount": 12,
+    "pageSha256": "52ee9cca70741f2efb77c8ca7a2b34d0fc485fe7f489dd6f8ce6866f7aca6a7f"
+  },
+  "registryFamilies": [
+    {
+      "id": "chat",
+      "name": "ch.at",
+      "tier": "A",
+      "protocols": [
+        "openai-chat",
+        "HTTP query",
+        "SSH",
+        "DNS"
+      ],
+      "anonymousAuth": "none",
+      "nativeTools": false,
+      "structuredOutput": null,
+      "context": {
+        "historyBytes": 65536
+      },
+      "assessment": "already_supported_text_only",
+      "primarySources": [
+        "https://github.com/Deep-ai-inc/ch.at"
+      ],
+      "notes": [
+        "Provider-selected model; omit model on the wire rather than inventing a supported model ID.",
+        "Parent verified a synthetic text stream; no duplicate generation probe."
+      ]
+    },
+    {
+      "id": "llm7",
+      "name": "LLM7",
+      "tier": "A",
+      "protocols": [
+        "openai-chat"
+      ],
+      "anonymousAuth": "universal Bearer unused placeholder",
+      "nativeTools": "per_model",
+      "structuredOutput": "per_model",
+      "context": {
+        "source": "model catalog context_window.tokens"
+      },
+      "assessment": "already_supported_native_chat",
+      "primarySources": [
+        "https://docs.llm7.io/quickstart",
+        "https://docs.llm7.io/guides/models"
+      ],
+      "notes": [
+        "Use explicit tools_calling/capabilities.tools and context_window.tokens metadata.",
+        "Exclude concrete usage_based_only:true models; free provider default selector is separate.",
+        "Parent owns full model compatibility matrix."
+      ]
+    },
+    {
+      "id": "ovh_ai_endpoints",
+      "name": "OVHcloud AI Endpoints",
+      "tier": "A",
+      "protocols": [
+        "openai-chat",
+        "openai-responses"
+      ],
+      "anonymousAuth": "none; omit Authorization",
+      "nativeTools": "per_model",
+      "structuredOutput": "per_model",
+      "context": {
+        "source": "dynamic public model metadata"
+      },
+      "assessment": "already_supported_native_chat",
+      "primarySources": [
+        "https://docs.ovhcloud.com/en/guides/public-cloud/ai-machine-learning/ai-endpoints-capabilities"
+      ],
+      "notes": [
+        "Native function/schema support depends on model.",
+        "Registry documents 2 requests/minute/IP/model and 2MB body limit.",
+        "Parent owns full model compatibility matrix.",
+        "Pinned documented anonymous no-auth tier is separate from authenticated catalog prices; do not reject OVH anonymous chat rows solely for nonzero displayed prices. No general pricing waiver applies to other hosts."
+      ]
+    },
+    {
+      "id": "ai_horde",
+      "name": "AI Horde",
+      "tier": "A",
+      "protocols": [
+        "async Kobold text-completion job"
+      ],
+      "anonymousAuth": "universal anonymous apikey 0000000000",
+      "nativeTools": false,
+      "structuredOutput": false,
+      "context": {
+        "source": "worker capacity; no single global token guarantee",
+        "requireContextFit": "reject"
+      },
+      "assessment": "adapter_possible_but_no_completed_reply_verified",
+      "primarySources": [
+        "https://github.com/Haidra-Org/AI-Horde/blob/main/README_integration.md",
+        "https://aihorde.net/api/swagger.json"
+      ],
+      "adapter": [
+        "Serialize chat as text without claiming native tool/schema support.",
+        "Discover worker context capacity, submit once, poll within deadline, cancel on abort or timeout.",
+        "Use context_fit:reject to avoid default silent truncation."
+      ],
+      "evidence": [
+        "/tmp/omnius-keyless-horde-probe-20261012.json",
+        "/tmp/omnius-keyless-horde-probe-60s-20261012.json"
+      ],
+      "notes": [
+        "First anonymous POST accepted202; queued at ~30s estimated remaining after12.5s; DELETE cancellation200.",
+        "Second authorized 60s-budget attempt rejected immediately HTTP403; no job ID or pending work.",
+        "No further jobs submitted."
+      ]
+    },
+    {
+      "id": "hf_public_gradio_spaces",
+      "name": "Hugging Face public Gradio Spaces",
+      "tier": "B",
+      "protocols": [
+        "Gradio named-function queue/SSE"
+      ],
+      "anonymousAuth": "none for genuinely public API; quota/gates vary per Space",
+      "nativeTools": "per_space_not_inferred",
+      "structuredOutput": "per_space_not_inferred",
+      "context": {
+        "source": "per-Space application/model; no global guarantee"
+      },
+      "assessment": "dedicated_adapter_needed",
+      "primarySources": [
+        "https://huggingface.co/docs/hub/en/spaces-api-endpoints",
+        "https://huggingface.co/spaces/Qwen/Qwen3-Demo",
+        "https://huggingface.co/spaces/Qwen/Qwen3-Demo/blob/main/app.py"
+      ],
+      "adapter": [
+        "Resolve owner-published public Space URL, api_name and documented input/output schemas.",
+        "Preserve session initialization, queue receipts, streaming and cancellation.",
+        "Do not treat all public Spaces as generic native tool providers or bypass ZeroGPU/account gates."
+      ],
+      "evidence": [
+        "/tmp/omnius-keyless-qwen-gradio-probe-20261012.json"
+      ],
+      "notes": [
+        "Official Qwen3 demo stage RUNNING and /gradio_api/info200 with add_message.",
+        "Synthetic toggle/add_message POST200 but SSE error404 Session not found; no generation verified.",
+        "Official QwQ preview Space PAUSED. Official SmolLM browser WebGPU demos require client/local model execution and were excluded."
+      ]
+    },
+    {
+      "id": "hf_public_dedicated_endpoint",
+      "name": "Hugging Face Public Inference Endpoints",
+      "tier": "B",
+      "protocols": [
+        "deployment-specific OpenAI/TGI/vLLM/custom"
+      ],
+      "anonymousAuth": "none only when endpoint owner deliberately selects Public access",
+      "nativeTools": "per_endpoint",
+      "structuredOutput": "per_endpoint",
+      "context": {
+        "source": "per-model deployment metadata"
+      },
+      "assessment": "no_authoritative_public_directory",
+      "primarySources": [
+        "https://huggingface.co/docs/inference-endpoints/security",
+        "https://huggingface.co/docs/inference-endpoints/guides/configuration"
+      ],
+      "notes": [
+        "Needs an explicitly owner-published endpoint URL and capability receipts.",
+        "Public access mode is not itself a discoverable hosted chat provider."
+      ]
+    },
+    {
+      "id": "inference_sh_embed",
+      "name": "inference.sh Embed API",
+      "tier": "B",
+      "protocols": [
+        "custom published-agent run/SSE/stop API"
+      ],
+      "anonymousAuth": "none for creator-published embed; optional creator cookies",
+      "nativeTools": "per_published_agent",
+      "structuredOutput": "per_published_agent",
+      "context": {
+        "source": "publisher/agent-specific"
+      },
+      "assessment": "not_generic_model_fallback",
+      "primarySources": [
+        "https://inference.sh/docs/api/rest/embed"
+      ],
+      "notes": [
+        "Requires exact publisher agent namespace/name; allowedOrigins can restrict callers.",
+        "Publisher funds usage; may return402 if its balance is unavailable.",
+        "Do not invent agent IDs or replace host tools/context with opaque hosted-agent behavior."
+      ]
+    },
+    {
+      "id": "jina_reader",
+      "name": "Jina Reader",
+      "tier": "B",
+      "protocols": [
+        "URL/PDF to extracted text HTTP"
+      ],
+      "anonymousAuth": "none for Reader",
+      "nativeTools": false,
+      "structuredOutput": false,
+      "context": {
+        "source": "document extraction limits; not conversational model context"
+      },
+      "assessment": "exclude_from_chat_fallback",
+      "primarySources": [
+        "https://github.com/jina-ai/reader",
+        "https://jina.ai/reader/"
+      ],
+      "notes": [
+        "Reader is an extraction tool, not a text/chat completion endpoint.",
+        "Do not confuse unauthenticated Reader with keyed Jina Search or model APIs."
+      ]
+    },
+    {
+      "id": "ahm7_makki",
+      "name": "AHM7 x Makki",
+      "tier": "B",
+      "protocols": [
+        "custom image-chat/OCR/TTS/image/audio REST"
+      ],
+      "anonymousAuth": "none per published docs",
+      "nativeTools": false,
+      "structuredOutput": null,
+      "context": {
+        "source": "not documented for general chat"
+      },
+      "assessment": "exclude_general_chat_requires_image",
+      "primarySources": [
+        "https://ahm7xmakki.com/docs"
+      ],
+      "notes": [
+        "Published /api/imgchat requires image:file and prompt:string.",
+        "No text-only general chat, native tool/schema or context contract documented.",
+        "Could support a separately scoped vision/tool adapter."
+      ]
+    },
+    {
+      "id": "pollinations_legacy",
+      "name": "Pollinations legacy anonymous hosts",
+      "tier": "C",
+      "protocols": [
+        "historical text HTTP/OpenAI-like; current gen API"
+      ],
+      "anonymousAuth": "current official API requires account API key",
+      "nativeTools": null,
+      "structuredOutput": null,
+      "context": {
+        "source": "model metadata only"
+      },
+      "assessment": "keep_disabled_not_currently_keyless",
+      "primarySources": [
+        "https://github.com/pollinations/pollinations"
+      ],
+      "notes": [
+        "Current official authentication docs require signup/API key.",
+        "Public /models GET does not establish anonymous generation availability; no generation probe."
+      ]
+    },
+    {
+      "id": "keylessai_relay",
+      "name": "KeylessAI community relay",
+      "tier": "C",
+      "protocols": [
+        "OpenAI-like community upstream relay"
+      ],
+      "anonymousAuth": "advertised none; host currently unresolved",
+      "nativeTools": null,
+      "structuredOutput": null,
+      "context": {
+        "source": "upstream-dependent"
+      },
+      "assessment": "disabled_unreachable_not_independent_capacity",
+      "primarySources": [
+        "https://github.com/lordbasilaiassistant-sudo/keylessai"
+      ],
+      "notes": [
+        "Advertised keylessai.thryx.workers.dev health host failed DNS resolution here.",
+        "Relay does not add independently owned model capacity."
+      ]
+    },
+    {
+      "id": "prexzy",
+      "name": "Prexzy APIs",
+      "tier": "C",
+      "protocols": [
+        "custom REST catalog; unknown upstreams"
+      ],
+      "anonymousAuth": "advertised keyless",
+      "nativeTools": null,
+      "structuredOutput": null,
+      "context": {
+        "source": "undocumented"
+      },
+      "assessment": "keep_disabled_no_verified_chat_contract",
+      "primarySources": [
+        "https://prexzyapis.com/"
+      ],
+      "notes": [
+        "Current public UI reports 0 active endpoints.",
+        "No directly verified text/chat protocol, tool/schema contract or functioning endpoint."
+      ]
+    }
+  ],
+  "additionalCandidates": [
+    {
+      "id": "opencode",
+      "baseUrl": "https://opencode.ai/inference/v1",
+      "catalogUrl": "https://models.opencode.ai/api.json",
+      "anonymousAuth": "universal public sentinel; user-authorized source-protocol compatibility identity",
+      "protocol": "openai-chat",
+      "assessment": "verified_native_and_safe_auxiliary_candidate",
+      "sources": [
+        "https://opencode.ai/docs/zen",
+        "https://github.com/anomalyco/opencode/tree/15b24ae12ddb8b7ae3e923b286b4032761bee510",
+        "https://models.opencode.ai/api.json"
+      ],
+      "sourceCommit": "15b24ae12ddb8b7ae3e923b286b4032761bee510",
+      "license": "MIT",
+      "metadataBytes": 5463169,
+      "metadataSha256": "53b0f8141a15c31a5a966bb95e37a35e49fe6c4345a60662daf25c1c6e94f832",
+      "freePredicate": "explicit numeric input=0 AND output=0; all advertised context tiers/cache charges also zero; nondeprecated text models; supported pinned OpenAI-compatible chat protocol",
+      "nativeProbe": {
+        "endpoint": "https://opencode.ai/inference/v1/chat/completions",
+        "model": "nemotron-3.5-lightning-free",
+        "synthetic": true,
+        "auth": "universal public sentinel",
+        "userAgent": "opencode/latest/2.0.27/cli",
+        "client": "cli",
+        "transparentClient": "omnius/1.0.847",
+        "sourceCommit": "15b24ae12ddb8b7ae3e923b286b4032761bee510",
+        "requestShape": "source-derived native-shaped, not binary-captured",
+        "systemPrefixSha256": "9876f53ad077e81d56ceb42ebd0313da8577d2947af44271faf973c454ad66fd",
+        "tools": [
+          "read",
+          "shell"
+        ],
+        "toolExecution": false,
+        "stream": true,
+        "requestBytes": 3838,
+        "status": 200,
+        "contentType": "text/event-stream",
+        "elapsedMs": 1044,
+        "visibleText": "OK",
+        "done": true,
+        "finishReasons": [
+          "stop",
+          "stop"
+        ],
+        "usage": {
+          "prompt_tokens": 1092,
+          "completion_tokens": 37,
+          "total_tokens": 1129,
+          "prompt_tokens_details": {
+            "audio_tokens": 0,
+            "cached_tokens": 0,
+            "cache_write_tokens": 0
+          },
+          "completion_tokens_details": {
+            "audio_tokens": 0,
+            "reasoning_tokens": 34
+          }
+        }
+      },
+      "schemaNoneProbe": {
+        "kind": "schema-none",
+        "endpoint": "https://opencode.ai/inference/v1/chat/completions",
+        "model": "nemotron-3.5-lightning-free",
+        "auth": "universal public sentinel",
+        "synthetic": true,
+        "toolExecution": false,
+        "requestBytes": 3997,
+        "toolChoice": "none",
+        "visibleText": "{\"value\":\"ok\"}",
+        "toolCalls": [],
+        "finishReasons": [
+          "stop",
+          "stop"
+        ],
+        "done": true,
+        "status": 200,
+        "contentType": "text/event-stream",
+        "usage": {
+          "prompt_tokens": 246,
+          "completion_tokens": 420,
+          "total_tokens": 666,
+          "prompt_tokens_details": {
+            "audio_tokens": 0,
+            "cached_tokens": 0,
+            "cache_write_tokens": 0
+          },
+          "completion_tokens_details": {
+            "audio_tokens": 0,
+            "reasoning_tokens": 413
+          }
+        },
+        "elapsedMs": 4120
+      },
+      "forcedCustomToolProbe": {
+        "kind": "forced-custom-tool",
+        "endpoint": "https://opencode.ai/inference/v1/chat/completions",
+        "model": "nemotron-3.5-lightning-free",
+        "auth": "universal public sentinel",
+        "synthetic": true,
+        "toolExecution": false,
+        "requestBytes": 4095,
+        "toolChoice": {
+          "type": "function",
+          "function": {
+            "name": "omnius_probe"
+          }
+        },
+        "visibleText": "",
+        "toolCalls": [
+          {
+            "name": "omnius_probe",
+            "arguments": "{\n  \"value\": \"ok\"\n}"
+          }
+        ],
+        "finishReasons": [
+          "tool_calls",
+          "tool_calls"
+        ],
+        "done": true,
+        "status": 200,
+        "contentType": "text/event-stream",
+        "usage": {
+          "prompt_tokens": 1177,
+          "completion_tokens": 32,
+          "total_tokens": 1209,
+          "prompt_tokens_details": {
+            "audio_tokens": 0,
+            "cached_tokens": 0,
+            "cache_write_tokens": 0
+          },
+          "completion_tokens_details": {
+            "audio_tokens": 0,
+            "reasoning_tokens": 21
+          }
+        },
+        "elapsedMs": 946
+      },
+      "negativeProbes": [
+        "/tmp/omnius-opencode-anonymous-probe-20261012.json",
+        "/tmp/omnius-opencode-compatibility-probe-20261012.json",
+        "/tmp/omnius-opencode-exact-compatibility-probe-20261012.json",
+        "/tmp/omnius-opencode-legacy-compatibility-probe-20261012.json"
+      ],
+      "notes": [
+        "Tool-less header-only probes returned403; source-derived native-shaped streamed request succeeded.",
+        "No server guard source or TLS fingerprint/signature requirement established.",
+        "No tool execution, private conversation, account credentials or local inference involved.",
+        "Metadata discovery alone remains catalog_only; runtime enrollment and live compatibility are separate receipts."
+      ],
+      "availability": {
+        "url": "https://opencode.ai/inference/v1/models",
+        "authorization": "omitted",
+        "status": 200,
+        "count": 83,
+        "bytes": 6967,
+        "sha256": "1d54eee7aaa8a6e5a97f9b41391ef12c21cf353045bbcb4321f676b9117794b3",
+        "eligibleMetadataIntersectionIds": [
+          "big-pickle",
+          "ling-3.1-flash-free",
+          "longcat-2.5-preview-free",
+          "mimo-v2.6-flash-free",
+          "nemotron-3.5-lightning-free",
+          "space-bunny-free",
+          "step-5-preview-free"
+        ]
+      }
+    },
+    {
+      "id": "llmfaucet",
+      "baseUrl": "https://api.llmfaucet.dev/v1",
+      "assessment": "unreachable_not_verified",
+      "sources": [
+        "https://github.com/llmfaucet/llmfaucet"
+      ],
+      "sourceCommit": "66434ae73f76d4992fc81d46411fd82b46e8431f",
+      "anonymousAuth": "advertised optional; universal free placeholder",
+      "notes": [
+        "Advertised API hostname fails DNS resolution here; no generation attempted.",
+        "README marks chat endpoint In progress and models endpoint Planned.",
+        "Its listed upstreams are already surveyed; no independent capacity established."
+      ]
+    },
+    {
+      "id": "airforce",
+      "assessment": "excluded_account_key_required",
+      "sources": [
+        "https://api.airforce/docs/api/chat/"
+      ],
+      "notes": [
+        "Primary docs require Bearer API key generated via dashboard; no generation probe."
+      ]
+    },
+    {
+      "id": "hackclub",
+      "assessment": "excluded_account_key_required",
+      "sources": [
+        "https://docs.ai.hackclub.com/guide/authentication.html"
+      ],
+      "notes": [
+        "Primary docs require API key for all requests; no generation probe."
+      ]
+    },
+    {
+      "id": "freeinference_org",
+      "assessment": "excluded_account_key_required",
+      "sources": [
+        "https://doc.freeinference.org/"
+      ],
+      "notes": [
+        "Primary setup docs require registration and API key; no generation probe."
+      ]
+    },
+    {
+      "id": "puter",
+      "assessment": "excluded_anonymous_CLI_OpenAI_gateway",
+      "sources": [
+        "https://developer.puter.com/tutorials/puter-auth-token/",
+        "https://docs.puter.com/rate-limits-and-quotas/"
+      ],
+      "notes": [
+        "Browser no-key API uses authenticated user-pays account/session flow.",
+        "Official CLI/backend flow requires account auth token; OpenAI/Anthropic-compatible endpoints require paid plan.",
+        "Anonymous browser budgets do not establish a keyless CLI model endpoint; no generation probe."
+      ]
+    },
+    {
+      "id": "kilo",
+      "baseUrl": "https://api.kilo.ai/api/gateway",
+      "catalogUrl": "https://api.kilo.ai/api/gateway/models",
+      "anonymousAuth": "none; omit Authorization",
+      "protocol": "openai-chat",
+      "assessment": "verified_text_native_tool_and_strict_schema",
+      "primarySources": [
+        "https://kilo.ai/docs/gateway/authentication",
+        "https://kilo.ai/docs/gateway/api-reference",
+        "https://kilo.ai/docs/gateway/models-and-providers",
+        "https://kilo.ai/landing/free-models"
+      ],
+      "rateLimitRequestsPerHourPerIp": 200,
+      "catalogBytes": 486989,
+      "catalogSha256": "5fa77a837d02c12c0a3b68f7fd7bf18c09c4f988c30a25de0a012311fe6330fd",
+      "catalogModelCount": 391,
+      "zeroInputOutputModels": [
+        {
+          "id": "kilo-auto/free",
+          "name": "Auto Free",
+          "context_length": 256000,
+          "pricing": {
+            "prompt": "0",
+            "completion": "0",
+            "input_cache_read": "0",
+            "input_cache_write": "0",
+            "request": "0",
+            "image": "0",
+            "web_search": "0",
+            "internal_reasoning": "0"
+          },
+          "supported_parameters": [
+            "max_tokens",
+            "temperature",
+            "tools",
+            "reasoning",
+            "include_reasoning"
+          ],
+          "architecture": {
+            "input_modalities": [
+              "text"
+            ],
+            "output_modalities": [
+              "text"
+            ],
+            "tokenizer": "Other"
+          },
+          "top_provider": {
+            "is_moderated": false,
+            "context_length": 256000,
+            "max_completion_tokens": 32768
+          },
+          "capabilities": null
+        },
+        {
+          "id": "inclusionai/ling-3.1-flash",
+          "name": "inclusionAI: Ling 3.1 Flash",
+          "context_length": 262144,
+          "pricing": {
+            "prompt": "0",
+            "completion": "0",
+            "discount": 0
+          },
+          "supported_parameters": [
+            "frequency_penalty",
+            "include_reasoning",
+            "logprobs",
+            "max_tokens",
+            "presence_penalty",
+            "reasoning",
+            "repetition_penalty",
+            "seed",
+            "stop",
+            "temperature",
+            "tool_choice",
+            "tools",
+            "top_k",
+            "top_logprobs",
+            "top_p"
+          ],
+          "architecture": {
+            "modality": "text->text",
+            "input_modalities": [
+              "text"
+            ],
+            "output_modalities": [
+              "text"
+            ],
+            "tokenizer": "Other"
+          },
+          "top_provider": {
+            "is_moderated": false,
+            "context_length": 262144,
+            "max_completion_tokens": 235929
+          },
+          "capabilities": null
+        },
+        {
+          "id": "dots-studio/dots-3-note-preview:free",
+          "name": "Dots Studio: Dots3-Note Preview (free)",
+          "context_length": 512000,
+          "pricing": {
+            "prompt": "0",
+            "completion": "0",
+            "discount": 0
+          },
+          "supported_parameters": [
+            "include_reasoning",
+            "max_tokens",
+            "reasoning",
+            "response_format",
+            "structured_outputs",
+            "temperature",
+            "tool_choice",
+            "tools",
+            "top_p"
+          ],
+          "architecture": {
+            "modality": "text+image->text",
+            "input_modalities": [
+              "text",
+              "image"
+            ],
+            "output_modalities": [
+              "text"
+            ],
+            "tokenizer": "Other"
+          },
+          "top_provider": {
+            "is_moderated": false,
+            "context_length": 512000,
+            "max_completion_tokens": 460800
+          },
+          "capabilities": null
+        },
+        {
+          "id": "liquid/lfm-2.5-2.6b:free",
+          "name": "LiquidAI: LFM2.5-2.6B (free)",
+          "context_length": 65536,
+          "pricing": {
+            "prompt": "0",
+            "completion": "0",
+            "discount": 0
+          },
+          "supported_parameters": [
+            "frequency_penalty",
+            "include_reasoning",
+            "logprobs",
+            "max_completion_tokens",
+            "max_tokens",
+            "min_p",
+            "presence_penalty",
+            "reasoning",
+            "repetition_penalty",
+            "response_format",
+            "seed",
+            "stop",
+            "structured_outputs",
+            "temperature",
+            "tool_choice",
+            "tools",
+            "top_k",
+            "top_logprobs",
+            "top_p"
+          ],
+          "architecture": {
+            "modality": "text->text",
+            "input_modalities": [
+              "text"
+            ],
+            "output_modalities": [
+              "text"
+            ],
+            "tokenizer": "Other"
+          },
+          "top_provider": {
+            "is_moderated": false,
+            "context_length": 65536,
+            "max_completion_tokens": 8192
+          },
+          "capabilities": null
+        },
+        {
+          "id": "nvidia/nemotron-3.5-lightning:free",
+          "name": "NVIDIA: Nemotron 3.5 Lightning (free)",
+          "context_length": 1000000,
+          "pricing": {
+            "prompt": "0",
+            "completion": "0",
+            "discount": 0
+          },
+          "supported_parameters": [
+            "include_reasoning",
+            "max_tokens",
+            "reasoning",
+            "seed",
+            "temperature",
+            "tool_choice",
+            "tools",
+            "top_p"
+          ],
+          "architecture": {
+            "modality": "text->text",
+            "input_modalities": [
+              "text"
+            ],
+            "output_modalities": [
+              "text"
+            ],
+            "tokenizer": "Other"
+          },
+          "top_provider": {
+            "is_moderated": false,
+            "context_length": 1000000,
+            "max_completion_tokens": 65536
+          },
+          "capabilities": null
+        },
+        {
+          "id": "thinkingmachines/inkling-small:free",
+          "name": "Thinking Machines: Inkling Small (free)",
+          "context_length": 1048576,
+          "pricing": {
+            "prompt": "0",
+            "completion": "0",
+            "discount": 0
+          },
+          "supported_parameters": [
+            "frequency_penalty",
+            "include_reasoning",
+            "max_tokens",
+            "presence_penalty",
+            "reasoning",
+            "reasoning_effort",
+            "seed",
+            "stop",
+            "temperature",
+            "tools",
+            "top_p"
+          ],
+          "architecture": {
+            "modality": "text+image+audio->text",
+            "input_modalities": [
+              "text",
+              "image",
+              "audio"
+            ],
+            "output_modalities": [
+              "text"
+            ],
+            "tokenizer": "Other"
+          },
+          "top_provider": {
+            "is_moderated": false,
+            "context_length": 1048576,
+            "max_completion_tokens": 262144
+          },
+          "capabilities": null
+        },
+        {
+          "id": "poolside/laguna-s-2.1:free",
+          "name": "Poolside: Laguna S 2.1 (free)",
+          "context_length": 262144,
+          "pricing": {
+            "prompt": "0",
+            "completion": "0",
+            "discount": 0
+          },
+          "supported_parameters": [
+            "include_reasoning",
+            "max_tokens",
+            "reasoning",
+            "temperature",
+            "tool_choice",
+            "tools"
+          ],
+          "architecture": {
+            "modality": "text->text",
+            "input_modalities": [
+              "text"
+            ],
+            "output_modalities": [
+              "text"
+            ],
+            "tokenizer": "Other"
+          },
+          "top_provider": {
+            "is_moderated": false,
+            "context_length": 262144,
+            "max_completion_tokens": 32768
+          },
+          "capabilities": null
+        },
+        {
+          "id": "poolside/laguna-xs-2.1:free",
+          "name": "Poolside: Laguna XS 2.1 (free)",
+          "context_length": 262144,
+          "pricing": {
+            "prompt": "0",
+            "completion": "0",
+            "discount": 0
+          },
+          "supported_parameters": [
+            "include_reasoning",
+            "max_tokens",
+            "reasoning",
+            "temperature",
+            "tool_choice",
+            "tools"
+          ],
+          "architecture": {
+            "modality": "text->text",
+            "input_modalities": [
+              "text"
+            ],
+            "output_modalities": [
+              "text"
+            ],
+            "tokenizer": "Other"
+          },
+          "top_provider": {
+            "is_moderated": false,
+            "context_length": 262144,
+            "max_completion_tokens": 32768
+          },
+          "capabilities": null
+        },
+        {
+          "id": "cohere/north-mini-code:free",
+          "name": "Cohere: North Mini Code (free)",
+          "context_length": 256000,
+          "pricing": {
+            "prompt": "0",
+            "completion": "0",
+            "discount": 0
+          },
+          "supported_parameters": [
+            "frequency_penalty",
+            "include_reasoning",
+            "max_tokens",
+            "presence_penalty",
+            "reasoning",
+            "seed",
+            "stop",
+            "temperature",
+            "tool_choice",
+            "tools",
+            "top_k",
+            "top_p"
+          ],
+          "architecture": {
+            "modality": "text->text",
+            "input_modalities": [
+              "text"
+            ],
+            "output_modalities": [
+              "text"
+            ],
+            "tokenizer": "Cohere"
+          },
+          "top_provider": {
+            "is_moderated": true,
+            "context_length": 256000,
+            "max_completion_tokens": 64000
+          },
+          "capabilities": null
+        },
+        {
+          "id": "nvidia/nemotron-3.5-content-safety:free",
+          "name": "NVIDIA: Nemotron 3.5 Content Safety (free)",
+          "context_length": 128000,
+          "pricing": {
+            "prompt": "0",
+            "completion": "0",
+            "discount": 0
+          },
+          "supported_parameters": [
+            "include_reasoning",
+            "max_tokens",
+            "reasoning",
+            "seed",
+            "temperature",
+            "top_p"
+          ],
+          "architecture": {
+            "modality": "text+image->text",
+            "input_modalities": [
+              "text",
+              "image"
+            ],
+            "output_modalities": [
+              "text"
+            ],
+            "tokenizer": "Other"
+          },
+          "top_provider": {
+            "is_moderated": false,
+            "context_length": 128000,
+            "max_completion_tokens": 8192
+          },
+          "capabilities": null
+        },
+        {
+          "id": "nvidia/nemotron-3-ultra-550b-a55b:free",
+          "name": "NVIDIA: Nemotron 3 Ultra (free)",
+          "context_length": 1000000,
+          "pricing": {
+            "prompt": "0",
+            "completion": "0",
+            "discount": 0
+          },
+          "supported_parameters": [
+            "include_reasoning",
+            "max_tokens",
+            "reasoning",
+            "reasoning_effort",
+            "seed",
+            "temperature",
+            "tool_choice",
+            "tools",
+            "top_p"
+          ],
+          "architecture": {
+            "modality": "text->text",
+            "input_modalities": [
+              "text"
+            ],
+            "output_modalities": [
+              "text"
+            ],
+            "tokenizer": "Other"
+          },
+          "top_provider": {
+            "is_moderated": false,
+            "context_length": 1000000,
+            "max_completion_tokens": 65536
+          },
+          "capabilities": null
+        },
+        {
+          "id": "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free",
+          "name": "NVIDIA: Nemotron 3 Nano Omni (free)",
+          "context_length": 256000,
+          "pricing": {
+            "prompt": "0",
+            "completion": "0",
+            "discount": 0
+          },
+          "supported_parameters": [
+            "include_reasoning",
+            "max_tokens",
+            "reasoning",
+            "seed",
+            "temperature",
+            "tool_choice",
+            "tools",
+            "top_p"
+          ],
+          "architecture": {
+            "modality": "text+image+audio+video->text",
+            "input_modalities": [
+              "text",
+              "audio",
+              "image",
+              "video"
+            ],
+            "output_modalities": [
+              "text"
+            ],
+            "tokenizer": "Other"
+          },
+          "top_provider": {
+            "is_moderated": false,
+            "context_length": 256000,
+            "max_completion_tokens": 65536
+          },
+          "capabilities": null
+        },
+        {
+          "id": "google/lyria-3-pro-preview",
+          "name": "Google: Lyria 3 Pro Preview",
+          "context_length": 1048576,
+          "pricing": {
+            "prompt": "0",
+            "completion": "0",
+            "discount": 0
+          },
+          "supported_parameters": [
+            "max_tokens",
+            "response_format",
+            "seed",
+            "temperature",
+            "top_p"
+          ],
+          "architecture": {
+            "modality": "text+image->text+audio",
+            "input_modalities": [
+              "text",
+              "image"
+            ],
+            "output_modalities": [
+              "text",
+              "audio"
+            ],
+            "tokenizer": "Other"
+          },
+          "top_provider": {
+            "is_moderated": false,
+            "context_length": 1048576,
+            "max_completion_tokens": 65536
+          },
+          "capabilities": null
+        },
+        {
+          "id": "google/lyria-3-clip-preview",
+          "name": "Google: Lyria 3 Clip Preview",
+          "context_length": 1048576,
+          "pricing": {
+            "prompt": "0",
+            "completion": "0",
+            "discount": 0
+          },
+          "supported_parameters": [
+            "max_tokens",
+            "response_format",
+            "seed",
+            "temperature",
+            "top_p"
+          ],
+          "architecture": {
+            "modality": "text+image->text+audio",
+            "input_modalities": [
+              "text",
+              "image"
+            ],
+            "output_modalities": [
+              "text",
+              "audio"
+            ],
+            "tokenizer": "Other"
+          },
+          "top_provider": {
+            "is_moderated": false,
+            "context_length": 1048576,
+            "max_completion_tokens": 65536
+          },
+          "capabilities": null
+        },
+        {
+          "id": "nvidia/nemotron-3-super-120b-a12b:free",
+          "name": "NVIDIA: Nemotron 3 Super (free)",
+          "context_length": 262144,
+          "pricing": {
+            "prompt": "0",
+            "completion": "0",
+            "discount": 0
+          },
+          "supported_parameters": [
+            "include_reasoning",
+            "max_tokens",
+            "reasoning",
+            "reasoning_effort",
+            "response_format",
+            "seed",
+            "structured_outputs",
+            "temperature",
+            "tool_choice",
+            "tools",
+            "top_p"
+          ],
+          "architecture": {
+            "modality": "text->text",
+            "input_modalities": [
+              "text"
+            ],
+            "output_modalities": [
+              "text"
+            ],
+            "tokenizer": "Other"
+          },
+          "top_provider": {
+            "is_moderated": false,
+            "context_length": 262144,
+            "max_completion_tokens": 235929
+          },
+          "capabilities": null
+        },
+        {
+          "id": "openrouter/free",
+          "name": "OpenRouter Free Models Router",
+          "context_length": 200000,
+          "pricing": {
+            "prompt": "0",
+            "completion": "0"
+          },
+          "supported_parameters": [
+            "frequency_penalty",
+            "include_reasoning",
+            "logprobs",
+            "max_completion_tokens",
+            "max_tokens",
+            "min_p",
+            "presence_penalty",
+            "reasoning",
+            "reasoning_effort",
+            "repetition_penalty",
+            "response_format",
+            "seed",
+            "stop",
+            "structured_outputs",
+            "temperature",
+            "tool_choice",
+            "tools",
+            "top_k",
+            "top_logprobs",
+            "top_p"
+          ],
+          "architecture": {
+            "modality": "text+image->text",
+            "input_modalities": [
+              "text",
+              "image"
+            ],
+            "output_modalities": [
+              "text"
+            ],
+            "tokenizer": "Router"
+          },
+          "top_provider": {
+            "is_moderated": false,
+            "context_length": null,
+            "max_completion_tokens": null
+          },
+          "capabilities": null
+        }
+      ],
+      "textProbe": {
+        "url": "https://api.kilo.ai/api/gateway/chat/completions",
+        "model": "liquid/lfm-2.5-2.6b:free",
+        "authorization": "omitted",
+        "synthetic": true,
+        "toolExecution": false,
+        "visibleText": "OK",
+        "done": true,
+        "finishReasons": [
+          "stop",
+          "stop"
+        ],
+        "status": 200,
+        "contentType": "text/event-stream",
+        "responseBytes": 5893,
+        "usage": {
+          "prompt_tokens": 14,
+          "completion_tokens": 33,
+          "total_tokens": 47,
+          "cost": 0,
+          "is_byok": false,
+          "prompt_tokens_details": {
+            "cached_tokens": 0,
+            "cache_write_tokens": 0,
+            "audio_tokens": 0,
+            "video_tokens": 0
+          },
+          "cost_details": {
+            "upstream_inference_cost": 0,
+            "upstream_inference_prompt_cost": 0,
+            "upstream_inference_completions_cost": 0
+          },
+          "completion_tokens_details": {
+            "reasoning_tokens": 30,
+            "image_tokens": 0,
+            "audio_tokens": 0
+          }
+        },
+        "elapsedMs": 1358
+      },
+      "nativeToolProbe": {
+        "kind": "forced-tool",
+        "url": "https://api.kilo.ai/api/gateway/chat/completions",
+        "model": "nvidia/nemotron-3-super-120b-a12b:free",
+        "authorization": "omitted",
+        "synthetic": true,
+        "toolExecution": false,
+        "visibleText": "",
+        "toolCalls": [
+          {
+            "name": "omnius_probe",
+            "arguments": "{\"value\":\"ok\"}"
+          }
+        ],
+        "finishReasons": [
+          "tool_calls",
+          "tool_calls"
+        ],
+        "done": true,
+        "status": 200,
+        "contentType": "text/event-stream",
+        "responseBytes": 4760,
+        "usage": {
+          "prompt_tokens": 301,
+          "completion_tokens": 44,
+          "total_tokens": 345,
+          "cost": 0,
+          "is_byok": false,
+          "prompt_tokens_details": {
+            "cached_tokens": 0,
+            "cache_write_tokens": 0,
+            "audio_tokens": 0,
+            "video_tokens": 0
+          },
+          "cost_details": {
+            "upstream_inference_cost": 0,
+            "upstream_inference_prompt_cost": 0,
+            "upstream_inference_completions_cost": 0
+          },
+          "completion_tokens_details": {
+            "reasoning_tokens": 19,
+            "image_tokens": 0,
+            "audio_tokens": 0
+          }
+        },
+        "elapsedMs": 1140
+      },
+      "strictSchemaProbe": {
+        "kind": "schema",
+        "url": "https://api.kilo.ai/api/gateway/chat/completions",
+        "model": "nvidia/nemotron-3-super-120b-a12b:free",
+        "authorization": "omitted",
+        "synthetic": true,
+        "toolExecution": false,
+        "visibleText": "{\"value\": \"ok\"}",
+        "toolCalls": [],
+        "finishReasons": [
+          "stop",
+          "stop"
+        ],
+        "done": true,
+        "status": 200,
+        "contentType": "text/event-stream",
+        "responseBytes": 43003,
+        "usage": {
+          "prompt_tokens": 27,
+          "completion_tokens": 298,
+          "total_tokens": 325,
+          "cost": 0,
+          "is_byok": false,
+          "prompt_tokens_details": {
+            "cached_tokens": 0,
+            "cache_write_tokens": 0,
+            "audio_tokens": 0,
+            "video_tokens": 0
+          },
+          "cost_details": {
+            "upstream_inference_cost": 0,
+            "upstream_inference_prompt_cost": 0,
+            "upstream_inference_completions_cost": 0
+          },
+          "completion_tokens_details": {
+            "reasoning_tokens": 290,
+            "image_tokens": 0,
+            "audio_tokens": 0
+          }
+        },
+        "elapsedMs": 3641
+      },
+      "preferredVerifiedNativeModel": "nvidia/nemotron-3-super-120b-a12b:free",
+      "notes": [
+        "All three synthetic streamed requests omitted Authorization and completed with normal terminal receipts; no actions executed.",
+        "Strict eligibility requires current explicit zero input/output prices and a generic text-chat contract; public catalog access alone does not imply free generation.",
+        "Content-safety classifier and Lyria audio-generation rows should not enter general chat fallback.",
+        "NVIDIA trial/data-handling terms are documented by Kilo; all probes used only public synthetic prompts."
+      ]
+    }
+  ],
+  "verifiedNewNativeProviderIds": [
+    "opencode",
+    "kilo"
+  ],
+  "adapterCandidatesNotVerifiedUsable": [
+    "ai_horde",
+    "hf_public_gradio_spaces"
+  ],
+  "limitations": [
+    "No successful Horde generation; first accepted job canceled, second rejected403.",
+    "No successful Qwen generation; public named API reached but session initialization failed.",
+    "Public source claims are separated from live generation receipts.",
+    "Parent owns full per-model compatibility matrix and production seed enrollment."
+  ]
+}