LiquidAI: d1 vs Nace.AI: Drex v2.0
A side-by-side comparison of LiquidAI: d1 ($0.07 / 1M tokens) and Nace.AI: Drex v2.0 ($0.21 / 1M tokens) using current USD prices and capabilities from the api14 catalog. Both run through the same OpenAI-compatible API.
Specifications
| LiquidAI: d1 | Nace.AI: Drex v2.0 | |
|---|---|---|
| Provider | LiquidAI | Nace.AI |
| Input | $0.07 / 1M tokens | $0.21 / 1M tokens |
| Cache read | $0.07 / 1M tokens | - |
| Context window | 65,536 | 131,072 |
| Max output | 58,982 | 117,964 |
| Input | text, image | text |
| Output | decisions | decisions |
| Tool calling | ||
| Structured output | ||
| Reasoning control | ||
| Added | Oct 1, 2026 | Oct 12, 2026 |
Key differences
- LiquidAI: d1 is 67% cheaper on input tokens.
- Nace.AI: Drex v2.0 reads 2.0× more context (131,072 tokens).
- A request with 1,000 input and 500 output tokens costs $0.000070 with LiquidAI: d1 and $0.000210 with Nace.AI: Drex v2.0.
In depth
LiquidAI: d1 is a decisions model from LiquidAI, released on 1 October 2026. Nace.AI: Drex v2.0 is a decisions model from Nace.AI, released on 12 October 2026. Both sit in the same category, meaning they are built to return a decision as their output rather than free form prose, and both are served through api14 alongside each other. The two providers are separate companies, so the choice between them is a choice between two vendors as much as between two model ids.
The clearest difference is the context window. Drex v2.0 accepts up to 131072 tokens, twice the 65536 tokens that d1 accepts, so it can hold roughly double the input in a single call. The other difference is input modality. d1 accepts both text and images, while Drex v2.0 accepts text only. On output the two are the same: both return decisions, and neither lists any supported parameters, so you cannot tune sampling behaviour or similar knobs through the request body on either one. The qualitative comparison marks d1 as the cheaper input side, while output pricing is not meaningfully different between them, so the input side is where d1 has a cost edge.
For practical selection, start with what you are feeding the model. If your input includes images, such as screenshots, scanned forms or diagrams that the model must read before deciding, d1 is the only one of the two that can take that input at all, so it is the default pick. If your input is long text, such as a large policy document, a lengthy transcript or a big batch of records that has to be considered together, Drex v2.0 is the better fit because its context window is twice as large and d1 would force you to truncate or split the work. If your input is short or moderate text with no images, both can handle it, and d1's cheaper input side makes it the reasonable first choice, with Drex v2.0 as the fallback when the prompt grows past what d1 can hold.
Calling either model works the same way because api14 exposes an OpenAI-compatible API. You point your existing OpenAI client at the api14 base URL, keep your usual request shape, and set the model field to the id you want. Use "liquid/d1" for the LiquidAI model and "nace-ai/drex-v2.0" for the Nace.AI model. Everything else in the request stays as it is, so switching between the two is a one line change to the model id, which makes it cheap to test both against your own inputs before committing. Since neither model lists supported parameters, keep your request body to the fields both accept and avoid sending tuning options that will not apply.
There is one case where Drex v2.0 wins even on text only work: when the task genuinely needs the larger window and splitting the input would lose context that matters to the decision. A single call over a long document can be more reliable than several calls over chunks whose decisions then have to be reconciled, and that is exactly where the doubled context earns its place. Conversely, if images are part of the input, d1 remains the only option regardless of length.
Switch with one string
Both models use the same endpoint and request shape. Change the model value from "liquid/d1" to "nace-ai/drex-v2.0" and prompts, streaming and tools keep working.
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api14.com/api/v1",
apiKey: process.env.API14_API_KEY,
});
const result = await client.post("/decisions", { body: {
"model": "nace-ai/drex-v2.0",
"state": "Customer: I was charged twice for my order and want my money back.",
"questions": {
"team": {
"type": "choice",
"instructions": "Which team should handle this message?",
"criteria": {
"billing": "Payments, refunds, invoices",
"support": "Product questions and bugs",
"sales": "Pricing and new purchases"
}
},
"urgent": {
"type": "noul",
"instructions": "Does this need a reply today?",
"criteria": {
"true": "Money, outage or legal risk",
"false": "Can wait"
}
}
}
} });
console.log(result.answers);
More comparisons
- LiquidAI: d1 Latest vs LiquidAI: d1-ultra Latest
- LiquidAI: d1 Latest vs Baseten: Joey 1
- LiquidAI: d1 Latest vs LiquidAI: d1.1
- LiquidAI: d1 Latest vs Morph: Reflex
- LiquidAI: d1-ultra Latest vs Baseten: Joey 1
- LiquidAI: d1-ultra Latest vs LiquidAI: d1.1
- LiquidAI: d1-ultra Latest vs Morph: Reflex
- Baseten: Joey 1 vs LiquidAI: d1.1