Is Liquid d1 the same product as TypeSafe Jev?

No. Jev is TypeSafe's hosted System One service. Liquid d1 is a separate product with its own weights, terms, and ops model unless explicitly documented as compatible HTTP.

Does Liquid d1 beat Jev on Decision Index?

The public snapshot in our metrics file scores Liquid d1; compare the facts table to hosted Jev. The d1-3B model card discusses multimodal inputs; treat Decision Index as text-oriented unless you run your own eval on your modalities.

Can I use the TypeSafe SDK with Liquid d1?

d1-3B modality preprocessing may differ from hosted Jev; adapter work belongs in your inference pipeline, not only in the Hugging Face model card.

Should I leave hosted Jev entirely?

Not necessarily. Teams often keep Jev for gates that need vendor support while experimenting with Liquid d1 on lower-risk paths. Run shadow mode until metrics prove parity.

madbid.lol · The paid leaderboard for AI builders. Products, agents and agencies buy their rank.

Liquid d1 as a Jev alternative

Last verified: 2026-10-12.How we rank

LiquidAI ships d1-3B on Hugging Face under the LFM 1.0 license with decision-oriented tuning. Independent Decision Index scores land near 40 in edition 0.3.1, below Jev but above several specialist checkpoints. d1 targets efficient inference on edge GPUs.

You still need to wrap inference in your System One compatible server if you want drop-in SDK parity with api.typesafe.ai.

jev.aitools.fyi is an independent directory. We are not TypeSafe, Cloudflare, OpenAI, Perplexity, or the makers named on this page unless a listing is marked official.

Catalog listing: Liquid d1-3B (brand-focused page; this article is the Jev comparison).

Liquid d1 vs Jev at a glance

Topic	Liquid d1	TypeSafe Jev	Source
Maker	Liquid AI	TypeSafe AI	Hugging Face: LiquidAI/d1-3B
First public release	d1-3B on Hugging Face	Hosted System One (ongoing)	Hugging Face: LiquidAI/d1-3B
License / access	LFM 1.0 (Hugging Face card)	Hosted API (weights not downloadable)	HF model card (LFM license)
Hosted price (input)	not published	$0.042 per M input tokens (vendor list, 2026-10-12)	TypeSafe models page (jev-1.13.0 price table)
API shape	Self-host inference; bring your HTTP contract	POST /v1/systemone	HF model card; Jev: TypeSafe API reference
Question types	Decision tuned LFM outputs	Choice, Score, Noul	HF model card
Inputs	Text state per model card	Structured state string	HF model card
Model sizes	d1-3B featured	Hosted Jev (closed)	HF weights card
Latency	not published	not published	Hugging Face: LiquidAI/d1-3B (no published p95 SLA cited)
Decision Index (public)	40.08 (Decision Index 0.3.1, snapshot 2026-10-10; independent harness, not vendor-only)	60.11 (Decision Index 0.3.1, snapshot 2026-10-10; independent harness, not vendor-only)	Decision Index 0.3.1 public snapshot
Data handling	not published	Processed under TypeSafe terms	Self-hosted data handling not published on a single vendor page
SDKs	Liquid inference examples; integrate with your gateway	typesafe_sdk, gateways, MCP ecosystem	HF model card
Adoption signal	10,843 HF downloads, lfm-1.0 (2026-10-12)	not published	Hugging Face model card

What Liquid d1 is

You still need to wrap inference in your System One compatible server if you want drop-in SDK parity with api.typesafe.ai.

Accuracy and benchmarks

Independent Decision Index edition 0.3.1 lists Liquid d1 at 40.08 in our imported snapshot.

Fit and tradeoffs

Choose Liquid d1 when your constraints match its deployment model: You want compact LFM weights for edge GPUs; Decision Index score beats smaller niche models in your slides; You can operate inference without hosted Jev.

Pricing and ops

d1-3B ships under the LFM 1.0 license; legal review and edge GPU sizing matter as much as headline $/M comparisons to hosted Jev.

Migrating from hosted Jev

Keep your question map and threshold logic stable. Swap client configuration to point at the new base URL or SDK, then run shadow traffic until disagreement rates fall below your risk budget.

                // Before: hosted Jev
const client = new TypeSafe({ apiKey: process.env.TYPESAFE_API_KEY });
await client.systemone({ model: "jev-latest", state, questions });

// After: self-hosted System One compatible server (Kev, decider, SemIf, etc.)
const client = new TypeSafe({ baseURL: "http://localhost:8000" });
await client.systemone({ model: "liquid-d1-3b", state, questions });

Ecosystem links

LiquidAI d1-3B model card covers LFM license terms and modality notes; catalog Liquid listings link related edge stacks.

Product-specific notes

d1-3B uses the LFM 1.0 license, not Apache; legal review should read the HF card before you redistribute weights. Vendor model cards cite their own benchmark slices; compare against your labels before you replace Jev in production.

Multimodal inputs increase preprocessing complexity relative to text-only System One calls. Liquid documents d1 as a decision-tuned LFM stack; confirm your inference code supports the modality mix you need. Smaller multimodal checkpoints can win on edge devices when latency matters more than maximum benchmark breadth.

Independent Decision Index scores for d1-3B are mid-pack in our snapshot; do not treat them as production guarantees.

When to choose Liquid d1

You want compact LFM weights for edge GPUs
Decision Index score beats smaller niche models in your slides
You can operate inference without hosted Jev

When to stay on Jev

You need full System One primitives with vendor support
You want gateway billing and large option sets without tuning
Apache or MIT licenses are required for weights

FAQ

Is Liquid d1 the same product as TypeSafe Jev?: No. Jev is TypeSafe's hosted System One service. Liquid d1 is a separate product with its own weights, terms, and ops model unless explicitly documented as compatible HTTP.
Does Liquid d1 beat Jev on Decision Index?: The public snapshot in our metrics file scores Liquid d1; compare the facts table to hosted Jev. The d1-3B model card discusses multimodal inputs; treat Decision Index as text-oriented unless you run your own eval on your modalities.
Can I use the TypeSafe SDK with Liquid d1?: d1-3B modality preprocessing may differ from hosted Jev; adapter work belongs in your inference pipeline, not only in the Hugging Face model card.
Should I leave hosted Jev entirely?: Not necessarily. Teams often keep Jev for gates that need vendor support while experimenting with Liquid d1 on lower-risk paths. Run shadow mode until metrics prove parity.

Other Jev alternatives

Sources

Last verified 2026-10-12. Spotted an error? Contact us.

All Jev alternatives What Jev is