{
  "listings": [
    {
      "slug": "decision-3",
      "name": "Decision 3.0",
      "company": "vLLM Semantic Router project",
      "grade": "B",
      "fit": "yes",
      "why": "choice with a probability per option on /v1/systemone through vLLM, Engine mode or d3_server.py; 54 ms median text on d3 and 18 ms on d3-mini (MI325X). First on the Decision Index text board at 66.3. Languages not stated.",
      "source": "https://huggingface.co/vllm-sr/d3",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/decision-3"
    },
    {
      "slug": "liquid-d1",
      "name": "Liquid d1",
      "company": "Liquid AI, Inc.",
      "grade": "C",
      "fit": "yes",
      "why": "POST /decisions/v1/systemone returns a choice, a probability per option and confidence; $0.02 per million input tokens with $2 free credit. Arize measured p50 0.20 s and Dataiku about 86 percent on routing. d1-2B's card lists 16 languages.",
      "source": "https://docs.liquid.ai/lfm/models/d1",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/liquid-d1"
    },
    {
      "slug": "microsoft-decision-1",
      "name": "Microsoft-Decision-1",
      "company": "Microsoft",
      "grade": "E",
      "fit": "yes",
      "why": "choice with a probability per option on Foundry's /providers/microsoft/v1/systemone or OpenRouter's Decisions API at $0.042 per million input tokens; OpenRouter shows 0.42 s latency and Dataiku about 86 percent. Microsoft's perturbation test found zero flips on reversed or shuffled options.",
      "source": "https://learn.microsoft.com/en-us/azure/foundry/foundry-models/how-to/use-foundry-models-microsoft-decision",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/microsoft-decision-1"
    },
    {
      "slug": "openai-decisions-api",
      "name": "OpenAI Decisions API",
      "company": "OpenAI OpCo, LLC",
      "grade": "A",
      "fit": "yes",
      "why": "choice with probabilities and confidence, a documented 'other' fallback, and answers that can come back as refusal; Arize measured p50 0.16 s and Dataiku 91 percent, the best in its test. GPT-6 Luna needs the Build tier ($5 of credit).",
      "source": "https://developers.openai.com/api/docs/guides/decisions",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/openai-decisions-api"
    },
    {
      "slug": "pplx-decider",
      "name": "pplx-decider",
      "company": "Perplexity AI, Inc.",
      "grade": "BB",
      "fit": "yes",
      "why": "choice with 1 to 255 options returns a probability per option and confidence on /v1/decisions in Jev's state-and-questions shape, with an 'other: null' option in the example; $0.02 per million, 10 requests a second per organisation by default. Dataiku about 86 percent. Card lists no languages.",
      "source": "https://docs.perplexity.ai/docs/decisions/quickstart",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/pplx-decider"
    },
    {
      "slug": "pplx-decider-flash",
      "name": "pplx-decider-flash",
      "company": "Perplexity AI, Inc.",
      "grade": "B",
      "fit": "yes",
      "why": "Same /v1/decisions shape at $0.01 per million with one question answered on its own; Decider for Mac runs it at 127.0.0.1 with no key for private text. Decision Index 54.81, 10.74 behind the 27B. Card lists no languages.",
      "source": "https://docs.perplexity.ai/docs/decisions/quickstart",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/pplx-decider-flash"
    },
    {
      "slug": "celeris-1-decision",
      "name": "Celeris-1 Decision",
      "company": "Marqo Inc",
      "grade": "D",
      "fit": "caveat",
      "why": "Choice answers carry a probability per option and a rescaled confidence, up to 512 options, on /v1/systemone or /v1/decisions. Workspace activation still runs a capacity waitlist and credit needs a card with a $5 minimum. No latency figure is published for the decision model.",
      "source": "https://docs.celeris.ai/decisions.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/celeris-1-decision"
    },
    {
      "slug": "cloudflare-clef",
      "name": "Clef",
      "company": "Cloudflare, Inc.",
      "grade": "B",
      "fit": "caveat",
      "why": "choice with a probability per option, 1 to 64 questions a call, 10,000 free neurons a day. Vendor median 209 ms (Clef) and 39 ms (Clef-flash); Arize measured p50 0.50 s on RAGTruth-length inputs. Dataiku had Clef about 86 percent, Clef-flash 77. Languages unstated.",
      "source": "https://developers.cloudflare.com/workers-ai/models/clef/",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/cloudflare-clef"
    },
    {
      "slug": "convai-laya",
      "name": "Laya",
      "company": "Convai Innovations Pvt. Ltd.",
      "grade": "BB",
      "fit": "caveat",
      "why": "laya-serve answers /v1/systemone in 33 ms on a T4 with a Router across 100+ languages, but the base checkpoints need fine-tuning: 0.425 on Banking77, weak past about 20 options and 37 percent on Hebrew intent (yuv.ai), Portuguese fell to 0.500 after 0.4.0 (issue 1453).",
      "source": "https://huggingface.co/convaiinnovations/laya",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/convai-laya"
    },
    {
      "slug": "decider",
      "name": "Decider",
      "company": "Mark Marosi (Mapika)",
      "grade": "BB",
      "fit": "caveat",
      "why": "/v1/systemone with 2 to 255 options and a probability per option, 50 ms median on CUDA and 133 ms on an M1 Pro. English only by the README's limits; issue 9's Chinese cases failed, and the 2B needs bucket-like option names.",
      "source": "https://github.com/Mapika/decider#readme",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/decider"
    },
    {
      "slug": "drex-max-v1",
      "name": "Drex Max v1",
      "company": "Nace AI, Inc.",
      "grade": "C",
      "fit": "caveat",
      "why": "choice returns a probability per option through /v1/systemone on the Python loader, Kapitsa, MLX or Nace's llama.cpp fork. The Open RAIL-M licence bars commercial use above $1 million in revenue or funding, and the card lists no languages.",
      "source": "https://huggingface.co/nace-ai/drex-max-v1",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/drex-max-v1"
    },
    {
      "slug": "elephant-1",
      "name": "elephant-1",
      "company": "Baseten Labs, Inc.",
      "grade": "D",
      "fit": "caveat",
      "why": "choice with a probability per option from one shared-prefix prefill, hosted on Baseten's /v1/systemone at $0.035 per million input tokens or locally through qev's DecisionEngine. The card's language is English only, and self-hosting needs OneJev at a pinned commit plus serve.patch.",
      "source": "https://huggingface.co/baselabs/elephant-1",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/elephant-1"
    },
    {
      "slug": "gliclass",
      "name": "GLiClass",
      "company": "Knowledgator Engineering Ltd.",
      "grade": "C",
      "fit": "caveat",
      "why": "One forward pass scores every label with a softmax (classification_type single-label) up to 128 labels, locally on CPU or GPU. No Jev shape (issue 47 asks for one), F1 only on the cards, and issue 12 reports every label scoring 0.99.",
      "source": "https://docs.knowledgator.com/docs/frameworks/gliclass/usage/",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/gliclass"
    },
    {
      "slug": "granite-switch",
      "name": "Granite Switch 4.1",
      "company": "International Business Machines Corporation",
      "grade": "B",
      "fit": "caveat",
      "why": "The decision aLoRA returns a softmax over single-token option labels, read through vLLM chat completions with logprob_token_ids and max_completion_tokens 1; the caller writes the prompt, tokenises labels and chunks past 128 options. Decision Index 46.72 on the 8B (IBM's run).",
      "source": "https://huggingface.co/ibm-granite/granite-switch-4.1-8b",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/granite-switch"
    },
    {
      "slug": "jaredpalmer-kev",
      "name": "Kev",
      "company": "Jared Palmer",
      "grade": "BB",
      "fit": "caveat",
      "why": "/v1/systemone with 1 to 255 options and a probability per option; Kev-4B answers six questions in 41.5 ms on an L40S and Arize had Kev 9B at 48 of 52 on routing. English only by the card, so non-English tickets are a blocker.",
      "source": "https://github.com/jaredpalmer/kev#readme",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/jaredpalmer-kev"
    },
    {
      "slug": "jev-27b-vl",
      "name": "JEV-27B-VL",
      "company": "AutoTrust AI Lab",
      "grade": "C",
      "fit": "caveat",
      "why": "POST /v1/decide takes one question a request with 2 to 256 options and returns a probability per option, about 264 ms a row on an RTX PRO 6000; CLINC150 with 150 intents zero-shot is on the card. Needs an 80 GB GPU and the card's vLLM development build.",
      "source": "https://huggingface.co/autotrust/JEV-27B-VL",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/jev-27b-vl"
    },
    {
      "slug": "joey-1",
      "name": "joey-1",
      "company": "Baseten Labs, Inc.",
      "grade": "C",
      "fit": "caveat",
      "why": "choice with a probability per option from one shared-prefix prefill, hosted on Baseten's /v1/systemone at $0.01 per million input tokens or locally through qev's DecisionEngine. The card's language is English only, and self-hosting needs OneJev at a pinned commit plus serve.patch.",
      "source": "https://huggingface.co/baselabs/joey-1",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/joey-1"
    },
    {
      "slug": "mammoth-1",
      "name": "mammoth-1",
      "company": "Baseten",
      "grade": "D",
      "fit": "caveat",
      "why": "choice with a probability per option from one shared-prefix prefill, hosted on Baseten's /v1/systemone at $0.10 per million input tokens or locally through qev's DecisionEngine. The card's language is English only, and self-hosting needs OneJev at a pinned commit plus serve.patch.",
      "source": "https://huggingface.co/baselabs/mammoth-1",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/mammoth-1"
    },
    {
      "slug": "nace-drex",
      "name": "Drex 1.5",
      "company": "Nace AI, Inc.",
      "grade": "BB",
      "fit": "caveat",
      "why": "choice returns a probability per option and confidence on /v1/systemone, but drex-v1.5 is a retired id: the OpenAPI lists it among ids answered by drex-max-v1.1 at Max's limits and $0.03 per million. The $5 signup credit needs a Google or GitHub sign-in or a card. No latency figure published.",
      "source": "https://console.nace.ai/openapi.json",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/nace-drex"
    },
    {
      "slug": "rightnow-system-zero",
      "name": "RightNow System Zero",
      "company": "RightNow AI, Inc.",
      "grade": "B",
      "fit": "caveat",
      "why": "boolean, choice and score answers carry a decision, confidence and ranked alternatives, but every answer says calibrated: false in this release, latency headers come with no guarantee, and starting credit has its own rate and spend limits.",
      "source": "https://www.rightnowai.co/docs/system-zero/decide",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/rightnow-system-zero"
    },
    {
      "slug": "scaledown",
      "name": "ScaleDown",
      "company": "Carbon ScaleDown, Inc.",
      "grade": "E",
      "fit": "caveat",
      "why": "POST /v1/scaledown mirrors the Jev shape with a probability per option summing to 1, but /classify caps a request at 20 labels (16 multi-label), the key comes from sales, and no latency is published.",
      "source": "https://docs.scaledown.ai/api-reference/decisions",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/scaledown"
    },
    {
      "slug": "sprag-spev",
      "name": "SPEV",
      "company": "Sprag AI, Inc.",
      "grade": "D",
      "fit": "caveat",
      "why": "choice with 2 to 255 options returns a probability per option and confidence on an OpenAI-style /v1/decisions ($0.09 per million text tokens), with 60 requests a minute per organisation by default, a 402 without a payment method, and no published latency. Card says multilingual without a language list.",
      "source": "https://sprag.ai/docs/decisions/api",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/sprag-spev"
    },
    {
      "slug": "sqwish-kettei",
      "name": "kettei",
      "company": "Sqwish Ltd, trading as Sqwish Labs",
      "grade": "C",
      "fit": "caveat",
      "why": "POST /v1/decide (or the Jev shape on /v1/systemone) returns a probability per outcome for up to 128 decisions about one context, with a 'none' outcome and an abstain_below policy worked through in the ticket-routing example. New accounts may start on a waitlist, and no latency figure is published.",
      "source": "https://docs.sqwish.ai/concepts",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/sqwish-kettei"
    },
    {
      "slug": "strands-decider",
      "name": "Strands Decider 2B",
      "company": "Amazon Web Services, Inc.",
      "grade": "B",
      "fit": "caveat",
      "why": "/v1/systemone locally at 115 ms median on an RTX 3090 and 153 ms on an M3 Pro, but AWS Builder found German recall 16 percent, a 55 percent first-answer preference flipping on swap, routing AUC 0.51, and Dataiku had it at 65 percent against 86 for the leaders.",
      "source": "https://github.com/strands-labs/strands-decider#readme",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/strands-decider"
    },
    {
      "slug": "typesafe-jev",
      "name": "Jev",
      "company": "TypeSafe AI, Inc.",
      "grade": "B",
      "fit": "caveat",
      "why": "choice up to 255 options with probabilities and confidence, a companion-noul pattern for none-of-these and 0.7/0.3 threshold guidance; OpenRouter's test had 59 of 60 intents at 0.194 s p50 and Arize p50 0.08 s. Early access runs behind a waitlist with no free tier found.",
      "source": "https://docs.typesafe.ai/api",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/typesafe-jev"
    },
    {
      "slug": "vela",
      "name": "Vela 2.0",
      "company": "vLLM Semantic Router project (with KR Labs)",
      "grade": "BB",
      "fit": "caveat",
      "why": "choice over 2 to 255 options with a distribution through vela2_serve.py's /v1/systemone; MASSIVE unseen-scenario intent 0.727 on the 4B and a Jev Decision Index of 31.63, below its Decision 2.0 base. The 0.3B runs on CPU.",
      "source": "https://huggingface.co/vllm-sr/Vela-2.0-4B",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/vela"
    }
  ],
  "marks": "yes: the docs show it does the job as the success test says. caveat: it does the job with a limit an agent should know. no: it doesn't do this job. unchecked: we couldn't read it.",
  "meta": {
    "attribution": "Anchor Terminal (https://www.anchorterminal.com)",
    "attributionRequired": true,
    "credit": "CC BY 4.0, credit Anchor Terminal and link the page or API address you took it from (https://www.anchorterminal.com/terms/#data)",
    "docs": "https://www.anchorterminal.com/docs/",
    "generatedAt": "2026-11-16",
    "license": "CC-BY-4.0",
    "method": "https://www.anchorterminal.com/benchmark/",
    "methodology": "0.8",
    "openapi": "https://www.anchorterminal.com/openapi.json",
    "preview": false,
    "run": "2026-10-01",
    "runLabel": "October 2026 research run",
    "terms": "https://www.anchorterminal.com/terms/#data"
  },
  "useCase": {
    "accounts": 12,
    "after": [
      "inference.llm.draft-reply",
      "support.assign-queue",
      "human-review"
    ],
    "area": "models",
    "asks": [
      "which team should handle this ticket",
      "is this billing, shipping, technical or spam",
      "which model tier does this prompt need",
      "what is the intent of this message"
    ],
    "before": [
      "email.read",
      "support.fetch-ticket",
      "chat.receive-message"
    ],
    "blockers": [
      "waitlist or no free tier for a hosted model",
      "label set over the model's option cap (Laya about 20, Jev 255)",
      "non-English text with an English-only checkpoint",
      "hosted endpoint behind a bot check that refuses the SDK's default user agent"
    ],
    "capabilities": [
      "inference.decision"
    ],
    "category": "decision-models",
    "categoryName": "Decision models",
    "checked": "2026-11-12",
    "companies": {
      "fit": 5,
      "fitWithLimit": 18,
      "tagged": 30
    },
    "competition": "crowded",
    "fits": [
      "decision-3",
      "liquid-d1",
      "microsoft-decision-1",
      "openai-decisions-api",
      "pplx-decider",
      "pplx-decider-flash"
    ],
    "fitsWithLimit": [
      "celeris-1-decision",
      "cloudflare-clef",
      "convai-laya",
      "decider",
      "drex-max-v1",
      "elephant-1",
      "gliclass",
      "granite-switch",
      "jaredpalmer-kev",
      "jev-27b-vl",
      "joey-1",
      "mammoth-1",
      "nace-drex",
      "rightnow-system-zero",
      "scaledown",
      "sprag-spev",
      "sqwish-kettei",
      "strands-decider",
      "typesafe-jev",
      "vela"
    ],
    "id": "decision-models.route-to-queue",
    "listings": {
      "caveat": 20,
      "no": 0,
      "notRead": 1,
      "tagged": 33,
      "unchecked": 0,
      "yes": 6
    },
    "name": "Route a message or request to a queue, team, intent or model tier",
    "status": "draft",
    "success": "One label from the caller's list with a probability per option, in one call, with low-probability answers routable to a human by threshold. Measured: under 500 ms hosted or under 1 s local.",
    "url": "https://www.anchorterminal.com/api/v1/use-cases/decision-models.route-to-queue.json",
    "weight": 22,
    "weightBasis": "12 of 41 accounts; the most common job in the HN threads, the AWS Builder posts and the Dataiku and yuv.ai tests"
  }
}
