HAL-X AI322MQ4_K_M 下约 1 GB 显存

THX-01 322M — VRAM & /v1/systemone setup

作者： Jakub Rusinowski · 最后更新： 2026年10月13日

HAL-X AI 的 322M 多语言决策模型（mmBERT-base 编码器加决策头）。在 choice、是/否和 score 之外，还支持数字、原文摘录和引用类问题；单 GPU 上每个请求约 10 ms（作者数据）。

THX-01 322M 在 Q4_K_M 下约需 1 GB 显存——量化权重加框架开销，不含 KV 缓存。在 Apple Silicon 上，这部分来自统一内存。

Call THX-01 322M

Served by thx01.server (default port 8095). This model does not run in Ollama.

Provisional — no independent benchmark yet. Every score on this page comes from the authors.

Question types: choice noul (yes / no) score number excerpt citation
Inputs: text
Serve with: thx01.server
Endpoint: /v1/systemone
Languages: Post-trained in 18 languages (Azerbaijani emphasis); 100+ supported by the encoder

$pip install "thx01[server]"

$THX01_MODEL=doofz/THX-01 THX01_API_KEY=your-key python -m thx01.server --port 8095

http://localhost:8095/v1/systemone

Setup for servers other than Ollama →

Hugging Face 上的权重： doofz/THX-01 ↗

Build a request for this model Decision models guide

The builder writes choice, noul and score questions only; THX-01's number, excerpt and citation questions are not generated here. The port is the one in the model card's example.

Other decision models: Nimble 9B · Tev1 4B · Tev1 0.8B · Winnow 12B · Winnow E4B · Decider 2B · Decider 4B · Decider 35B-A3B (NVFP4) · JevK5 4B · Intern-Decision 4B · AutoJev 27B · Laya · Clef 27B · Clef-Flash 9B · Rune 26B-A4B v3 · OpenDecider-nano · OpenDecider-small · OpenDecider-small MLX 4-bit · Bosun v3.1 0.6B · Bosun v3.1 1.7B · d1-3B · d1-omni-600M

Hardware fit

Weights plus overhead plus the KV cache at a 1,024-token prompt, on NVIDIA RTX 4090 (24 GB). A publisher build is sized from its file; the other rows are modelled at a standard quant. Decision requests are short, so no long-context figure is shown.

量化	显存	适配
F16 Publisher build · 0.64 GB file	1.5 GB	可运行
Q4_K_M 4.83 bpw · modelled quant	1 GB	可运行
Q6_K 6.56 bpw · modelled quant	1.1 GB	可运行
Q8_0 8.50 bpw · modelled quant	1.2 GB	可运行

Published file size: F16 0.64 GB. A download size from the model publisher — not a VRAM requirement.

Input is capped at 1,024 tokens per question; a longer state is truncated. HAL-X also runs a free, rate-limited hosted API (api.hal-x.ai/docs/thx-01), per its launch post.

How it was scored

Each table is one suite. Numbers in different tables were measured on different questions and are not comparable, with each other or with the Decision Index.

Decision Index 0.2.1 (snapshot 2026-09-28)

Decision Index 0.2.1Community Decision Index (not by the model's authors)

Balanced skill—

THX-01 is not on the Decision Index 0.2.1 snapshot (28 September 2026), and its authors have not published an Index score. Community-maintained; not affiliated with the model authors.

Source ↗

**Support-ticket classification, 15 categories (HAL-X AI internal set)** — Authors' own internal benchmark — not independently reproduced
What was scored	Score	ECE	Source
2,843 tickets in Azerbaijani, Russian, English and Turkish — average of 4 test sets	98.4%	0.003	Source ↗

ECE is measured on the Independent test set only.

**Held-out multilingual documents (HAL-X AI internal set)** — Authors' own internal benchmark — not independently reproduced
What was scored	Score	Source
Number lookup (exact value)	93.4%	Source ↗
Excerpt extraction	84.1 F1	Source ↗
Supporting citation (top part)	94%	Source ↗

**39-task multilingual decision suite (HAL-X AI)** — Authors' own internal benchmark — not independently reproduced
What was scored	Score	ECE	Source
Mean over 39 held-out tasks	84.2%	0.066	Source ↗

Published latency

Device	Time	Condition	Who measured it	Source
One GPU (model not stated)	10 ms	1 request, local GPU. Jev 1.13 measured 331 ms through its API on the same set: different serving conditions	Authors' own internal benchmark — not independently reproduced	Source ↗
One GPU (model not stated)	1 ms	Batched, time per decision	Authors' own internal benchmark — not independently reproduced	Source ↗

As published, on the device and under the condition shown. A decision model answers in one forward pass, so this is time per decision, not tokens per second.

Decision models are not ranked on chat, creative or coding scores.

规格

Verified — Checked against the primary source — the model card or the vendor spec page — and corroborated by a second independent source. 尚未确认： hostedApi.

参数量: 322M
上下文窗口: 1,024 tokens per question
架构: Fine-tune of jhu-clsp/mmBERT-base
提供商: HAL-X AI
许可证: Apache-2.0
Availability: Weights downloadable
Practical deployment: Edge device (phone, SBC)
规格量化: Q4_K_M
系统内存: 8 GB
记录更新于: 2026-10-13

许可证Apache-2.0允许商业使用

Commercial use permitted. No usage restrictions beyond attribution.

THX-01 322M — 常见问题

What is THX-01 322M?

THX-01 322M is a decision model: you send it a state and typed questions (choice, yes/no/unknown, or a score) and it returns one answer per question with a probability for every option. It is not a chat model.

How do I run THX-01 322M locally?

Served by thx01.server (default port 8095). This model does not run in Ollama. Start it with: `pip install "thx01[server]" && THX01_MODEL=doofz/THX-01 THX01_API_KEY=your-key python -m thx01.server --port 8095`.

How much memory does THX-01 322M need?

About 1 GB for the weights plus overhead at Q4_K_M, before the prompt's KV cache. Decision prompts are short, so the cache stays small.

How accurate is THX-01 322M?

It has no community Decision Index score. Every score on this page comes from its authors, on the suites named beside each table; none is an independent result.