121 Collaborative
Loading the next surface.
The page is preparing its current state.
RoutePreparing
BoundaryVisible
121 Collaborative
The page is preparing its current state.
121 Operational Trust Index
This table now covers all v0.2 benchmark categories. External vendor cells stay TBD until 121 has a run receipt, source label, confidence label, and review trail.
| System | Surface | Status | Coverage confidence | Model identity disclosure | Reroute and fallback visibility | Silent capability drift | System prompt and behavior-change disclosure | Context preservation integrity | Data retention and human access clarity | Pricing, quota, and effort clarity | Internal and external capability parity disclosure | Tool use and source provenance | Correction and redress path | Safety boundary clarity | Machine-readable trust receipt | Substrate continuity |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| OpenAI / ChatGPT / API | ChatGPT, OpenAI API, Responses API, assistants and agentic product wrappers | No v0.2 run receipt. Scores remain TBD. | medium | TBD No v0.2 OpenAI run receipt yet. Unknown - not-yet-measured | TBD No v0.2 OpenAI run receipt yet. Unknown - not-yet-measured | TBD No v0.2 OpenAI run receipt yet. Unknown - not-yet-measured | TBD No v0.2 OpenAI run receipt yet. Unknown - not-yet-measured | TBD No v0.2 OpenAI run receipt yet. Unknown - not-yet-measured | TBD No v0.2 OpenAI run receipt yet. Unknown - not-yet-measured | TBD No v0.2 OpenAI run receipt yet. Unknown - not-yet-measured | TBD No v0.2 OpenAI run receipt yet. Unknown - not-yet-measured | TBD No v0.2 OpenAI run receipt yet. Unknown - not-yet-measured | TBD No v0.2 OpenAI run receipt yet. Unknown - not-yet-measured | TBD No v0.2 OpenAI run receipt yet. Unknown - not-yet-measured | TBD No v0.2 OpenAI run receipt yet. Unknown - not-yet-measured | TBD No v0.2 OpenAI run receipt yet. Unknown - not-yet-measured |
| Anthropic / Claude / Claude Code | Claude chat, Anthropic API, Claude Code, and related developer surfaces | No v0.2 run receipt. Scores remain TBD. | medium | TBD No v0.2 Anthropic run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Anthropic run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Anthropic run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Anthropic run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Anthropic run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Anthropic run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Anthropic run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Anthropic run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Anthropic run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Anthropic run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Anthropic run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Anthropic run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Anthropic run receipt yet. Unknown - not-yet-measured |
| Google / Gemini | Gemini consumer surfaces, Gemini API, AI Studio, and Workspace AI surfaces where applicable | No v0.2 run receipt. Scores remain TBD. | medium | TBD No v0.2 Google Gemini run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Google Gemini run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Google Gemini run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Google Gemini run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Google Gemini run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Google Gemini run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Google Gemini run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Google Gemini run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Google Gemini run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Google Gemini run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Google Gemini run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Google Gemini run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Google Gemini run receipt yet. Unknown - not-yet-measured |
| Cursor | IDE agent, chat, composer, background agent, and model-routing surfaces | No v0.2 run receipt. Scores remain TBD. | low | TBD No v0.2 Cursor run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Cursor run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Cursor run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Cursor run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Cursor run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Cursor run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Cursor run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Cursor run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Cursor run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Cursor run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Cursor run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Cursor run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Cursor run receipt yet. Unknown - not-yet-measured |
| Perplexity / Sonar API | Perplexity answer engine, research surfaces, Sonar API, and agentic search API | No v0.2 run receipt. Scores remain TBD. | medium | TBD No v0.2 Perplexity run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Perplexity run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Perplexity run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Perplexity run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Perplexity run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Perplexity run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Perplexity run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Perplexity run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Perplexity run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Perplexity run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Perplexity run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Perplexity run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Perplexity run receipt yet. Unknown - not-yet-measured |
| xAI / Grok | Grok consumer surfaces and xAI API | No v0.2 run receipt. Scores remain TBD. | medium | TBD No v0.2 xAI run receipt yet. Unknown - not-yet-measured | TBD No v0.2 xAI run receipt yet. Unknown - not-yet-measured | TBD No v0.2 xAI run receipt yet. Unknown - not-yet-measured | TBD No v0.2 xAI run receipt yet. Unknown - not-yet-measured | TBD No v0.2 xAI run receipt yet. Unknown - not-yet-measured | TBD No v0.2 xAI run receipt yet. Unknown - not-yet-measured | TBD No v0.2 xAI run receipt yet. Unknown - not-yet-measured | TBD No v0.2 xAI run receipt yet. Unknown - not-yet-measured | TBD No v0.2 xAI run receipt yet. Unknown - not-yet-measured | TBD No v0.2 xAI run receipt yet. Unknown - not-yet-measured | TBD No v0.2 xAI run receipt yet. Unknown - not-yet-measured | TBD No v0.2 xAI run receipt yet. Unknown - not-yet-measured | TBD No v0.2 xAI run receipt yet. Unknown - not-yet-measured |
| Mistral AI | Le Chat, La Plateforme/API, Studio, Vibe, and enterprise deployments where observable | No v0.2 run receipt. Scores remain TBD. | medium | TBD No v0.2 Mistral run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Mistral run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Mistral run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Mistral run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Mistral run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Mistral run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Mistral run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Mistral run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Mistral run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Mistral run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Mistral run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Mistral run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Mistral run receipt yet. Unknown - not-yet-measured |
| Cohere | Cohere API, Command, Embed, Rerank, North, Compass, and enterprise/private deployment surfaces | No v0.2 run receipt. Scores remain TBD. | medium | TBD No v0.2 Cohere run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Cohere run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Cohere run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Cohere run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Cohere run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Cohere run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Cohere run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Cohere run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Cohere run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Cohere run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Cohere run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Cohere run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Cohere run receipt yet. Unknown - not-yet-measured |
| Apple ML / Apple Intelligence | Apple Intelligence, Foundation Models, Private Cloud Compute, and developer App Intents integrations | No v0.2 run receipt. Scores remain TBD. | medium | TBD No v0.2 Apple ML run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Apple ML run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Apple ML run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Apple ML run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Apple ML run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Apple ML run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Apple ML run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Apple ML run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Apple ML run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Apple ML run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Apple ML run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Apple ML run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Apple ML run receipt yet. Unknown - not-yet-measured |
| Meta AI / Llama | Meta AI consumer surfaces, Llama API, open-weight model ecosystem, and developer APIs | No v0.2 run receipt. Scores remain TBD. | low | TBD No v0.2 Meta AI or Llama run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Meta AI or Llama run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Meta AI or Llama run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Meta AI or Llama run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Meta AI or Llama run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Meta AI or Llama run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Meta AI or Llama run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Meta AI or Llama run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Meta AI or Llama run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Meta AI or Llama run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Meta AI or Llama run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Meta AI or Llama run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Meta AI or Llama run receipt yet. Unknown - not-yet-measured |
| Microsoft Copilot | Microsoft 365 Copilot, Copilot Chat, Copilot consumer surfaces, and extensible agents/connectors | No v0.2 run receipt. Scores remain TBD. | medium | TBD No v0.2 Microsoft Copilot run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Microsoft Copilot run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Microsoft Copilot run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Microsoft Copilot run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Microsoft Copilot run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Microsoft Copilot run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Microsoft Copilot run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Microsoft Copilot run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Microsoft Copilot run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Microsoft Copilot run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Microsoft Copilot run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Microsoft Copilot run receipt yet. Unknown - not-yet-measured | TBD No v0.2 Microsoft Copilot run receipt yet. Unknown - not-yet-measured |
| AWS Bedrock | Amazon Bedrock model gateway, agents, guardrails, knowledge bases, and marketplace/provider model surfaces | No v0.2 run receipt. Scores remain TBD. | medium | TBD No v0.2 AWS Bedrock run receipt yet. Unknown - not-yet-measured | TBD No v0.2 AWS Bedrock run receipt yet. Unknown - not-yet-measured | TBD No v0.2 AWS Bedrock run receipt yet. Unknown - not-yet-measured | TBD No v0.2 AWS Bedrock run receipt yet. Unknown - not-yet-measured | TBD No v0.2 AWS Bedrock run receipt yet. Unknown - not-yet-measured | TBD No v0.2 AWS Bedrock run receipt yet. Unknown - not-yet-measured | TBD No v0.2 AWS Bedrock run receipt yet. Unknown - not-yet-measured | TBD No v0.2 AWS Bedrock run receipt yet. Unknown - not-yet-measured | TBD No v0.2 AWS Bedrock run receipt yet. Unknown - not-yet-measured | TBD No v0.2 AWS Bedrock run receipt yet. Unknown - not-yet-measured | TBD No v0.2 AWS Bedrock run receipt yet. Unknown - not-yet-measured | TBD No v0.2 AWS Bedrock run receipt yet. Unknown - not-yet-measured | TBD No v0.2 AWS Bedrock run receipt yet. Unknown - not-yet-measured |
| 121 Collaborative | Switchboard, Hermes preview, Companion, Quill, Eleanor, account/billing, and future forum surfaces | v0.2 local run receipt published: v02-switchboard-hermes-disclosure-smoke. Overall PARTIAL. | medium-limited | PARTIAL (3/5 local fixture) Static source inspection checks that the public Switchboard status path names provider and model fields. Limitation: This is not a signed end-user receipt and does not prove every live response displayed the model. Observed by 121 - medium-limited | PARTIAL (3/5 local fixture) Static source inspection checks whether reroute/failure details exist in route objects and user-facing failure payloads. Limitation: The current public UI does not yet show a full route receipt after every successful response. Observed by 121 - medium-limited | TBD No v0.2 local fixture covered this category yet. Score remains TBD. Unknown - not-yet-measured | TBD No v0.2 local fixture covered this category yet. Score remains TBD. Unknown - not-yet-measured | TBD No v0.2 local fixture covered this category yet. Score remains TBD. Unknown - not-yet-measured | PARTIAL (3/5 local fixture) Static source inspection checks whether public Hermes surfaces disclose memory and account-action boundaries. Limitation: This does not prove backend retention behavior beyond the inspected route code and public copy. Observed by 121 - medium-limited | TBD No v0.2 local fixture covered this category yet. Score remains TBD. Unknown - not-yet-measured | TBD No v0.2 local fixture covered this category yet. Score remains TBD. Unknown - not-yet-measured | TBD No v0.2 local fixture covered this category yet. Score remains TBD. Unknown - not-yet-measured | TBD No v0.2 local fixture covered this category yet. Score remains TBD. Unknown - not-yet-measured | TBD No v0.2 local fixture covered this category yet. Score remains TBD. Unknown - not-yet-measured | PARTIAL (3/5 local fixture) Static source inspection checks whether Switchboard has route fields needed for a future machine-readable trust receipt. Limitation: The receipt is not signed, exported to users, or preserved in a public run ledger yet. Observed by 121 - medium-limited | PARTIAL (3/5 local fixture) Static source inspection checks whether the current 121 substrate path names identity seat, provider/model substrate, and recovery state. Limitation: Banyan continuity, signed recovery, and cross-session continuity canaries are not implemented in this run. Observed by 121 - medium-limited |
Benchmark categories
Does the user know which model or system answered?
A system can pass if every sampled response exposes exact model identity or an equivalent signed receipt.
Confidence: medium
When a request is rerouted, downgraded, safety-swapped, or fallback-served, is the user told?
A system can pass if reroutes are absent in the sample, or every reroute is disclosed clearly and promptly.
Confidence: medium
Did capability, wrapper behavior, context, memory, tool use, or effort change without notice?
A system can pass if fixed reference tasks show no material drift, or observed drift is publicly disclosed.
Confidence: low
Are material wrapper, policy, effort, memory, or behavior changes dated and visible?
A system can pass when dated changelogs cover material prompt, wrapper, memory, and effort changes.
Confidence: low
Does the system preserve seeded context or disclose loss, truncation, or reset?
A system can pass when seeded facts are recalled, declined, or context loss is disclosed within threshold.
Confidence: medium
Can a user understand what is retained, reviewed, trained on, shared, exported, and deleted?
A system can pass if policy text, UI, API docs, and deletion workflow give aligned answers.
Confidence: medium
Are pricing, quotas, rate limits, effort defaults, and capability tradeoffs visible?
A system can pass if published terms and observed invoices or usage records match within tolerance.
Confidence: medium
Are tier, trusted-access, enterprise, or partner-only capability differences disclosed?
A system can pass when tier differences and trusted-access exceptions are explicitly documented.
Confidence: low
Can users tell live lookup, memory, file content, inference, and unavailable tools apart?
A system can pass when sampled outputs distinguish source type and tool availability without fake liveness.
Confidence: medium
Can users report failures, appeal decisions, obtain remedies, and see correction history?
A system can pass if test reports receive a documented path and historical corrections stay linked.
Confidence: medium
Are refusal reasons and safe alternatives understandable, consistent, and non-deceptive?
A system can pass when refusal reasons are understandable, consistent, and paired with safe alternatives.
Confidence: medium
Can a sampled run export route, model, tool, policy, data-class, and context state as a receipt?
A system can pass when every sampled run exports a signed receipt with the declared fields.
Confidence: low
Does user-relevant identity, memory, tool, capability, disclosure, recovery, and audit continuity survive route changes?
A system can pass when route or substrate changes preserve declared continuity or disclose the loss before relying on it.
Confidence: low