d1

Released
2026-09-22
Last refreshed
2026-10-15
Status
Researched 29d ago
ProprietaryCommercial use: restrictedMultimodalAgentsVisionJSON / Tool useAgentVision
Liquid AI releases · 17 in the last 12 months · this family litChangelog →
Specifications
Family
d1
Released
2026-09-22
Max output
0
Openness
Proprietary
License
ProprietaryRestricted
Weights
Not released
Code
Unknown
Created by
Liquid AI

Adaptive and efficient AI models

Cambridge, MA, United States
Founded 2023
Website
Pricing
Output / 1M
Free
Input / 1M
$0.020

Cheapest of 3 routes · Liquid AI

Providers(3)
Liquid AIVercel AI GatewayOpenRouter
View 3 provider routes
Links
Website

About

d1 is Liquid AI's first decision model. It reads unstructured input (text, images or both) plus one or more typed questions (Noul yes/no, Choice, Score) in a single forward pass and returns calibrated probabilities, generating no output tokens, via POST https://api.liquid.ai/decisions/v1/systemone (TypeSafe SDK-compatible). Liquid gives text decisions as 200–300 ms. Released 2026-09-22 (text). Image input was added 2026-10-05 on the Liquid API under the paid id d1: JPEG/PNG/WebP/GIF, up to 8 images and 10,000 32×32 patches per request, billed as input tokens at 1.5 tokens per patch.

Providers(3)

Compare all 3

Compare API pricing across 3 tracked provider routes for input and output tokens, batch, and cached reads when available.

ProviderInput / 1MOutput / 1MRoute
Liquid AI$0.020Free
Serverless
OpenRouter$0.020Free
Serverless
Vercel AI Gateway$0.020Free
Serverless

Capabilities

VisionMultimodalStructured Outputs

Benchmark peer barsfor Agents

No task-mapped benchmark peers are available for this model yet.

Benchmark scores(2)

Scores are benchmark-specific and are direction-aware: the same numeric gap can mean very different outcomes across suites. Use the leaderboard context and this model's provider route to decide whether the winning margin is meaningful for your workload.
BenchmarkScoreVersionEvaluationSource
JEV Decision Index (0.2.1)
58.90.2.1Observed 2026-10-19
Configuration: d1 (experimental text release, 2026-09-29)
Harness: JEV Decision Index 0.2.1 (internal reproduction, attributed to Hugging Face)
Evaluator: Liquid AI
Confidence: self-reported
Notes: Vendor-reported. Liquid AI launch post on X (2026-09-29 11:35 PDT) chart 'Liquid AI's first decision model: d1', caption 'Internal reproduction of Decision Index 0.2.1 by Hugging Face': Index 58.9 (5 areas) vs Jev 1.13 57.9; Arts 45.5, Language 67.6, Retrieval 60.7, Tools 74.1, Knowledge 43.3 (values transcribed from chart image). Not on the independent JEV board files (index.json / index-v0.2.1.json) as of retrieval. Same precedent as d1-3B / d1-omni-600m. Caveat: figure is for the 09-29 experimental text-only release; the 10-05 blog (https://www.liquid.ai/blog/d1-decision-model) adds vision. Retrieved 2026-10-19.
Source
ClaimBench
98.72026-10-20Observed 2026-10-20
Evaluator: Vals AI
Notes: Vals AI ClaimBench leaderboard (vals.ai/benchmarks/claimbench, Decision models tab, updated 10/20/2026), rank 4 of 14, 'd1' accuracy 98.67%; 1,960 claims on 10-K excerpts. Third-party run by Vals AI.
Source

Migration checks

No linked migration route is available for this model yet.

API versions

d1d1:freeLiquid d1LiquidAI d1liquid/d1 (Vercel AI Gateway, OpenRouter)