# Dia / Dia2 Review

> Dia / Dia2 review: Best-known open model for scripted two-speaker dialogue rather than narration - Apache-2.0, English-only, GPU-oriented. Verified pricing.

![Dia / Dia2 logo](https://www.versusref.com/logos/dia.png)

Best-known open model for scripted two-speaker dialogue rather than narration - Apache-2.0, English-only, GPU-oriented.

Among the 46 text-to-speech tools we track, Dia / Dia2 has the 43rd-widest language coverage.

See pricing · Open source · facts verified Jul 20, 2026

## What we know about Dia / Dia2

Dia / Dia2 sits in the text-to-speech apis category, where it is best-known open model for scripted two-speaker dialogue rather than narration - Apache-2.0, English-only, GPU-oriented. We keep this Dia / Dia2 profile grounded in primary sources, each fact dated to when we last confirmed it.

Dia / Dia2 does not publish a public per-minute rate we have been able to verify, so the price figure here stays blank until we can confirm one. We would rather show nothing than a guessed number; if you have a current Dia / Dia2 quote, it is the fastest way for us to close that gap.

On capabilities, Dia / Dia2 covers streaming audio output, instant voice cloning, emotion / style controls, and self-host / on-prem option. Each of those is verified against Dia / Dia2's own docs or dashboard, not marketing copy.

Dia / Dia2 ranks the 43rd-widest language coverage of the 46 text-to-speech tools we track, so where it lands for you depends on which of those matters more.

In total we track 12 verified facts for Dia / Dia2 today, each linking the primary source it came from so you can check our work - and vendor claims we have not measured ourselves are labeled as such on the Dia / Dia2 fact sheet below.

## Fact sheet

### Capabilities

| Fact | Value | Verified | Source |
| --- | --- | --- | --- |
| Streaming audio output | ✓  Yes | Jul 20 | [source](https://github.com/nari-labs/dia2) |
| Instant voice cloning | ✓  Yes | Jul 20 | [source](https://github.com/nari-labs/dia) |
| Languages supported | 1 languages | Jul 20 | [source](https://github.com/nari-labs/dia) |
| Emotion / style controls | ✓  Yes | Jul 20 | [source](https://github.com/nari-labs/dia) |

### Compliance & trust

| Fact | Value | Verified | Source |
| --- | --- | --- | --- |
| Self-host / on-prem option | ✓  Yes | Jul 20 | [source](https://github.com/nari-labs/dia) |
| Model weights license | Apache-2.0 | Jul 20 | [source](https://github.com/nari-labs/dia) |

### Build experience

| Fact | Value | Verified | Source |
| --- | --- | --- | --- |
| Official SDKs | PyTorch (pip) | Jul 20 | [source](https://github.com/nari-labs/dia2) |
| Max input per request | Dia2: up to ~2 minutes of generated audio (English) | Jul 20 | [source](https://github.com/nari-labs/dia2) |

### Commercial

| Fact | Value | Verified | Source |
| --- | --- | --- | --- |
| Model size (parameters) | 1.6B (Dia); 1B and 2B (Dia2) | Jul 20 | [source](https://github.com/nari-labs/dia2) |
| Hardware to self-host | Dia: GPU | Jul 20 | [source](https://github.com/nari-labs/dia) |
| Project maintenance status | dormant | Jul 20 | [source](https://github.com/nari-labs/dia/commits/main) |
| GitHub stars | 19,300 stars | Jul 20 | [source](https://github.com/nari-labs/dia) |

## Dia / Dia2 head-to-head

| Comparison | Record | Per use case |
| --- | --- | --- |
| [Dia / Dia2 vs ElevenLabs](https://www.versusref.com/tts/dia-vs-elevenlabs/) | won 1 · lost 5 · tied 0 | Audiobooks: lost; Content Creators: lost; Developers: lost; Dubbing: lost; Self-Hosted: won; Voice Agents: lost |

Source: https://www.versusref.com/tts/tools/dia/