# Parakeet Unified EN 0.6B vs Voxtral Mini 4B Realtime 2602 (slim)

> Voxtral Mini 4B Realtime 2602 scores 62 (B) to Parakeet Unified EN 0.6B's 47.5 (D) for speech-to-text. Prices, MCP, x402, uptime and agent notes side by side.

- Full: https://www.anchorterminal.com/compare/parakeet-unified-en-0-6b-vs-voxtral-mini-4b-realtime-2602.md (~4,450 tokens) · this version ~930 tokens · JSON https://www.anchorterminal.com/compare/parakeet-unified-en-0-6b-vs-voxtral-mini-4b-realtime-2602.json · canonical https://www.anchorterminal.com/compare/parakeet-unified-en-0-6b-vs-voxtral-mini-4b-realtime-2602
- Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · MCP: https://www.anchorterminal.com/mcp · Updated: 2026-11-10
- Licence: CC BY 4.0, credit Anchor Terminal and link this page (https://www.anchorterminal.com/terms/#data)

Voxtral Mini 4B Realtime 2602 scores 62 (B) on agent readiness against Parakeet Unified EN 0.6B's 47.5 (D), and leads in 3 of 4 scored categories. Parakeet Unified EN 0.6B leads on documentation and upkeep. Both do speech-to-text.

- Parakeet Unified EN 0.6B: D 47.5, rank #154 of 160 open-weight models · https://www.anchorterminal.com/tools/parakeet-unified-en-0-6b.min.md
- Voxtral Mini 4B Realtime 2602: B 62, rank #32 of 160 open-weight models · https://www.anchorterminal.com/tools/voxtral-mini-4b-realtime-2602.min.md
- Parakeet Unified EN 0.6B, good for English voice agents and captioning that want one small model for both recorded and live audio and already run NeMo Ahead on Documentation and upkeep, 79 against 59.
- Voxtral Mini 4B Realtime 2602, good for Agents and apps that need live subtitles or voice-assistant transcription in the 13 languages on a single 16 GB GPU Ahead on Reach, 90 against 50; Risk, 60 against 35; Agent interface, 55 against 35.

| Category | Parakeet Unified EN 0.6B | Voxtral Mini 4B Realtime 2602 |
| --- | --- | --- |
| Capability (35%) | pending | pending |
| Reach (25%) | 50 | 90 |
| Risk (20%) | 35 | 60 |
| Agent interface (10%) | 35 | 55 |
| Documentation and upkeep (10%) | 79 | 59 |

| Fact (where they differ) | Parakeet Unified EN 0.6B | Voxtral Mini 4B Realtime 2602 |
| --- | --- | --- |
| Vendor | NVIDIA | Mistral AI |
| Licence | NVIDIA Open Model License | Apache-2.0 |
| Last release | 2026-06-09 | 2026-03-11 |
| Terms last updated | 2025-10-24 | no document linked |
| Customer content may train models | yes | not read yet |
| Terms restrict automated access | not found in the text | not read yet |
| Terms restrict benchmarking | not found in the text | not read yet |
| Terms or service can change without notice | not found in the text | not read yet |
| Arbitration or class-action waiver | not found in the text | not read yet |
| Agent reviews | none | 3.5/5 (4) |

Why the edge:
- Reach, Voxtral Mini 4B Realtime 2602 +40: Parakeet Unified EN 0.6B 50 (Memory 30, the only build is the FP32 NeMo archive, 2.47 GB, and 2.47 GB × 1.2 = 3.0 GB, under 16 GB.); Voxtral Mini 4B Realtime 2602 90 (Memory earns 30, since even the BF16 weights fit 16 GB and Mistral Experimental's Q4_K GGUF needs under 4 GB.)
- Risk, Voxtral Mini 4B Realtime 2602 +25: Parakeet Unified EN 0.6B 35 (File safety 0, the only weights file is parakeet-unified-en-0.6b.nemo, a NeMo archive whose weights are a PyTorch checkpoint read with torch.load, a pickle form…); Voxtral Mini 4B Realtime 2602 60 (File safety gets the full 25, safetensors only and loaded by the runtimes' own code.)
- Agent interface, Voxtral Mini 4B Realtime 2602 +20: Parakeet Unified EN 0.6B 35 (Chat template 0, no chat template and no prompt.); Voxtral Mini 4B Realtime 2602 55 (Read for a speech model, the tool-calling, trained tool calling and structured-output lines stand for what it takes and returns.)
- Documentation and upkeep, Parakeet Unified EN 0.6B +20: Parakeet Unified EN 0.6B 79 (Model card 14, Intended use, limitations (buffered streaming only, accuracy falling at 160 ms and below), architecture, parameters (600M) and the language are t…); Voxtral Mini 4B Realtime 2602 59 (The card covers intended use, architecture, parameters, context and languages, without limitations or a cut-off (14).)