Rank 3 · model deployment
Google Gemini 3 Flash API is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
candidatecurrent2 evidence rowsPolicy tool-policy-shadow-2026-05-07
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 4 · model deployment
Google Gemini 2.5 Flash API is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
candidatecurrent2 evidence rowsPolicy tool-policy-shadow-2026-05-07
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 12 · model deployment
OpenAI GPT-5.5 API is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
candidatecurrent1 evidence rowsPolicy tool-policy-shadow-2026-05-07
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 13 · model deployment
Anthropic Claude Sonnet 4.6 API is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
candidatecurrent1 evidence rowsPolicy tool-policy-shadow-2026-05-07
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 23 · model deployment
Google Gemini 3 Pro API is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
candidatecurrent2 evidence rowsPolicy tool-policy-shadow-2026-05-07
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 25 · model deployment
Anthropic Claude Opus 4.7 API is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
candidatecurrent1 evidence rowsPolicy tool-policy-shadow-2026-05-07
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 8 · model deployment
Llama 3.1 via Azure AI Foundry is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
candidate3 evidence rowsPolicy tool-policy-shadow-2026-05-07
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 14 · model deployment
Cohere Command R+ is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
candidate3 evidence rowsPolicy tool-policy-shadow-2026-05-07
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 16 · model deployment
Azure OpenAI GPT-4.1 is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
candidate1 evidence rowsPolicy tool-policy-shadow-2026-05-07
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 26 · model deployment
Mistral via Azure is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
candidate3 evidence rowsPolicy tool-policy-shadow-2026-05-07
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 28 · model deployment
Mistral 8x7B is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
candidate3 evidence rowsPolicy tool-policy-shadow-2026-05-07
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 29 · model deployment
Mistral 7B Instruct is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
candidate3 evidence rowsPolicy tool-policy-shadow-2026-05-07
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 30 · model deployment
Mistral Small is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
candidate3 evidence rowsPolicy tool-policy-shadow-2026-05-07
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 33 · model deployment
OpenAI GPT-4.1 is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
candidate1 evidence rowsPolicy tool-policy-shadow-2026-05-07
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 36 · model deployment
GPT-4.1 via OpenAI is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
candidate1 evidence rowsPolicy tool-policy-shadow-2026-05-07
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 40 · model deployment
gpt-4.1 (OpenAI) is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
candidate1 evidence rowsPolicy tool-policy-shadow-2026-05-07
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 41 · model deployment
Llama 3 via AWS Bedrock is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
candidate3 evidence rowsPolicy tool-policy-shadow-2026-05-07
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 42 · model deployment
Llama 3 on Azure is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
candidate3 evidence rowsPolicy tool-policy-shadow-2026-05-07
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 43 · model deployment
Llama 3 via Azure is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
candidate3 evidence rowsPolicy tool-policy-shadow-2026-05-07
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 44 · model deployment
Llama 3 via Bedrock is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
candidate3 evidence rowsPolicy tool-policy-shadow-2026-05-07
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 45 · model deployment
Mistral Codestral is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
candidate3 evidence rowsPolicy tool-policy-shadow-2026-05-07
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 46 · model deployment
Mistral API is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
candidate3 evidence rowsPolicy tool-policy-shadow-2026-05-07
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 47 · model deployment
Mistral Large via Le Chat API is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
candidate3 evidence rowsPolicy tool-policy-shadow-2026-05-07
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 48 · model deployment
Mistral Large via API is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
candidate3 evidence rowsPolicy tool-policy-shadow-2026-05-07
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 49 · model deployment
Mistral AI API is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
candidate3 evidence rowsPolicy tool-policy-shadow-2026-05-07
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 50 · model deployment
Mistral Large via APIs is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
candidate3 evidence rowsPolicy tool-policy-shadow-2026-05-07
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 51 · model deployment
Mistral AI is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
candidate3 evidence rowsPolicy tool-policy-shadow-2026-05-07
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 1 · model deployment
GPT-4o via Azure OpenAI is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
deprecatedsuperseded3 evidence rowsPolicy tool-policy-shadow-2026-05-07
This option is lifecycle-blocked for public recommendations. Review OpenAI GPT-5.5 or GPT-5.4 mini as the replacement.
OpenAI GPT-4o is superseded; review OpenAI GPT-5.5 or GPT-5.4 mini.
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 2 · model deployment
Azure OpenAI (GPT‑4o) is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
deprecatedsuperseded3 evidence rowsPolicy tool-policy-shadow-2026-05-07
This option is lifecycle-blocked for public recommendations. Review OpenAI GPT-5.5 or GPT-5.4 mini as the replacement.
OpenAI GPT-4o is superseded; review OpenAI GPT-5.5 or GPT-5.4 mini.
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 5 · model deployment
OpenAI GPT-4o is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
deprecatedsuperseded3 evidence rowsPolicy tool-policy-shadow-2026-05-07
This option is lifecycle-blocked for public recommendations. Review OpenAI GPT-5.5 or GPT-5.4 mini as the replacement.
OpenAI GPT-4o is superseded; review OpenAI GPT-5.5 or GPT-5.4 mini.
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 6 · model deployment
OpenAI GPT‑4o API is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
deprecatedsuperseded3 evidence rowsPolicy tool-policy-shadow-2026-05-07
This option is lifecycle-blocked for public recommendations. Review OpenAI GPT-5.5 or GPT-5.4 mini as the replacement.
OpenAI GPT-4o is superseded; review OpenAI GPT-5.5 or GPT-5.4 mini.
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 7 · model deployment
OpenAI GPT-4o (via API) is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
deprecatedsuperseded3 evidence rowsPolicy tool-policy-shadow-2026-05-07
This option is lifecycle-blocked for public recommendations. Review OpenAI GPT-5.5 or GPT-5.4 mini as the replacement.
OpenAI GPT-4o is superseded; review OpenAI GPT-5.5 or GPT-5.4 mini.
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 9 · model deployment
Google Vertex Gemini is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
deprecateddeprecated3 evidence rowsPolicy tool-policy-shadow-2026-05-07
This option is lifecycle-blocked for public recommendations. Review Google Gemini 3 Pro or Gemini 2.5 Pro as the replacement.
Google Gemini 1.5 Pro is deprecated; use Google Gemini 3 Pro or Gemini 2.5 Pro.
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 10 · model deployment
Google Gemini 1.5 Flash is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
deprecateddeprecated3 evidence rowsPolicy tool-policy-shadow-2026-05-07
This option is lifecycle-blocked for public recommendations. Review Google Gemini 2.5 Flash or Gemini 2.5 Flash-Lite as the replacement.
Google Gemini 1.5 Flash is deprecated; use Google Gemini 2.5 Flash or Gemini 2.5 Flash-Lite.
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 11 · model deployment
Anthropic Claude 3.5 Opus is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
deprecateddeprecated3 evidence rowsPolicy tool-policy-shadow-2026-05-07
This option is lifecycle-blocked for public recommendations. Review Anthropic Claude Opus 4.7 as the replacement.
Anthropic Claude 3.5 Opus is deprecated; use Anthropic Claude Opus 4.7.
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 15 · model deployment
Anthropic Claude on AWS is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
deprecateddeprecated2 evidence rowsPolicy tool-policy-shadow-2026-05-07
This option is lifecycle-blocked for public recommendations. Review Anthropic Claude Sonnet 4.6 as the replacement.
Anthropic Claude 3.5 Sonnet is deprecated; use Anthropic Claude Sonnet 4.6.
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 17 · model deployment
Anthropic Claude via AWS is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
deprecateddeprecated2 evidence rowsPolicy tool-policy-shadow-2026-05-07
This option is lifecycle-blocked for public recommendations. Review Anthropic Claude Sonnet 4.6 as the replacement.
Anthropic Claude 3.5 Sonnet is deprecated; use Anthropic Claude Sonnet 4.6.
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 18 · model deployment
Anthropic Claude via AWS Bedrock is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
deprecateddeprecated2 evidence rowsPolicy tool-policy-shadow-2026-05-07
This option is lifecycle-blocked for public recommendations. Review Anthropic Claude Sonnet 4.6 as the replacement.
Anthropic Claude 3.5 Sonnet is deprecated; use Anthropic Claude Sonnet 4.6.
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 19 · model deployment
Anthropic Claude via Bedrock is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
deprecateddeprecated2 evidence rowsPolicy tool-policy-shadow-2026-05-07
This option is lifecycle-blocked for public recommendations. Review Anthropic Claude Sonnet 4.6 as the replacement.
Anthropic Claude 3.5 Sonnet is deprecated; use Anthropic Claude Sonnet 4.6.
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 20 · model deployment
Anthropic Claude 3 Opus is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
deprecateddeprecated3 evidence rowsPolicy tool-policy-shadow-2026-05-07
This option is lifecycle-blocked for public recommendations. Review Anthropic Claude Opus 4.7 as the replacement.
Anthropic Claude 3 Opus is deprecated; use Anthropic Claude Opus 4.7.
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 21 · model deployment
Google Gemini 1.5 Pro is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
deprecateddeprecated3 evidence rowsPolicy tool-policy-shadow-2026-05-07
This option is lifecycle-blocked for public recommendations. Review Google Gemini 3 Pro or Gemini 2.5 Pro as the replacement.
Google Gemini 1.5 Pro is deprecated; use Google Gemini 3 Pro or Gemini 2.5 Pro.
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 22 · model deployment
Google Gemini 1.5 is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
deprecateddeprecated3 evidence rowsPolicy tool-policy-shadow-2026-05-07
This option is lifecycle-blocked for public recommendations. Review Google Gemini 3 Pro or Gemini 2.5 Pro as the replacement.
Google Gemini 1.5 Pro is deprecated; use Google Gemini 3 Pro or Gemini 2.5 Pro.
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 24 · model deployment
OpenAI GPT-4o mini is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
deprecatedsuperseded2 evidence rowsPolicy tool-policy-shadow-2026-05-07
This option is lifecycle-blocked for public recommendations. Review OpenAI GPT-5.5, GPT-5.4 mini, or GPT-5.4 nano as the replacement.
OpenAI GPT-4o mini is superseded; review OpenAI GPT-5.5, GPT-5.4 mini, or GPT-5.4 nano.
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 27 · model deployment
Anthropic Claude 3 Haiku is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
deprecateddeprecated3 evidence rowsPolicy tool-policy-shadow-2026-05-07
This option is lifecycle-blocked for public recommendations. Review Anthropic Claude Haiku 4.5 as the replacement.
Anthropic Claude 3 Haiku is deprecated; use Anthropic Claude Haiku 4.5.
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 31 · model deployment
Anthropic Claude 3.5 is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
deprecateddeprecated2 evidence rowsPolicy tool-policy-shadow-2026-05-07
This option is lifecycle-blocked for public recommendations. Review Anthropic Claude Sonnet 4.6 as the replacement.
Anthropic Claude 3.5 Sonnet is deprecated; use Anthropic Claude Sonnet 4.6.
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 32 · model deployment
Anthropic Claude 3.5 Sonnet is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
deprecateddeprecated2 evidence rowsPolicy tool-policy-shadow-2026-05-07
This option is lifecycle-blocked for public recommendations. Review Anthropic Claude Sonnet 4.6 as the replacement.
Anthropic Claude 3.5 Sonnet is deprecated; use Anthropic Claude Sonnet 4.6.
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 34 · model deployment
Anthropic Claude 3 is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
deprecateddeprecated2 evidence rowsPolicy tool-policy-shadow-2026-05-07
This option is lifecycle-blocked for public recommendations. Review Anthropic Claude Sonnet 4.6 as the replacement.
Anthropic Claude 3.5 Sonnet is deprecated; use Anthropic Claude Sonnet 4.6.
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 35 · model deployment
Anthropic Claude 3 Sonnet is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
deprecateddeprecated2 evidence rowsPolicy tool-policy-shadow-2026-05-07
This option is lifecycle-blocked for public recommendations. Review Anthropic Claude Sonnet 4.6 as the replacement.
Anthropic Claude 3.5 Sonnet is deprecated; use Anthropic Claude Sonnet 4.6.
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 37 · model deployment
Anthropic Claude 3.5 Sonnet API is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
deprecateddeprecated2 evidence rowsPolicy tool-policy-shadow-2026-05-07
This option is lifecycle-blocked for public recommendations. Review Anthropic Claude Sonnet 4.6 as the replacement.
Anthropic Claude 3.5 Sonnet is deprecated; use Anthropic Claude Sonnet 4.6.
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 38 · model deployment
Claude 3.5 via Anthropic is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
deprecateddeprecated2 evidence rowsPolicy tool-policy-shadow-2026-05-07
This option is lifecycle-blocked for public recommendations. Review Anthropic Claude Sonnet 4.6 as the replacement.
Anthropic Claude 3.5 Sonnet is deprecated; use Anthropic Claude Sonnet 4.6.
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.
Rank 39 · model deployment
Anthropic Claude via API is a candidate policy-ranked Quality text generation option. Public recommendation authority still requires gates to promote this candidate.
deprecateddeprecated2 evidence rowsPolicy tool-policy-shadow-2026-05-07
This option is lifecycle-blocked for public recommendations. Review Anthropic Claude Sonnet 4.6 as the replacement.
Anthropic Claude 3.5 Sonnet is deprecated; use Anthropic Claude Sonnet 4.6.
Why it ranks here
Quality and fit scores are lane-specific, so this rank is not a global popularity score.
Lifecycle status must stay current enough for the candidate to be rendered as a safe public choice.
Evidence rows and policy state decide whether the rank is advisory, candidate, or public authority.
Generated as a candidate from a candidate component_offering. Not approved for public Recommended rendering until authority gates pass.