tommytracx commited on
Commit
a7ceb96
·
verified ·
1 Parent(s): 1eb567a

TXF-MP v1.0 presentation: hero, title lockup, badge table, highlights, specifications, platforms, citation (ThoxWork-Code-14B)

Browse files
Files changed (1) hide show
  1. README.md +64 -3
README.md CHANGED
@@ -18,14 +18,48 @@ tags:
18
  - ollama
19
  - configuration-derived
20
  ---
 
21
 
22
- # ThoxWork-Code-14B — THOX Work coding build (Qwen3-14B)
23
 
24
- ![License](https://img.shields.io/badge/License-Apache_2.0-05A451) ![Params](https://img.shields.io/badge/Params-14B-05A451) ![Base](https://img.shields.io/badge/Base-Qwen3-14B-05A451) ![Access](https://img.shields.io/badge/Access-Public-05A451)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
25
 
26
  > **Your AI. Your Data. Your Rules.™**
27
 
28
- ThoxWork-Code-14B is a THOX-configured model: upstream Qwen3-14B weights with a THOX system profile and the versioned THOX Knowledge Pack v0.1. This repo holds the packaging; the runnable model lives on Ollama (`ollama run Thox-ai/thoxwork-code-14b`). THOX did not train these weights.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
29
 
30
  ## Files
31
 
@@ -50,6 +84,13 @@ hf download Thox-ai/ThoxWork-Code-14B --local-dir ./ThoxWork-Code-14B
50
  cd ./ThoxWork-Code-14B && ollama create thoxwork-code-14b -f Modelfile && ollama run thoxwork-code-14b
51
  ```
52
 
 
 
 
 
 
 
 
53
  ## Summary
54
  THOX Work build: the coding system profile, the enterprise workgroup prompt pack and the THOX Knowledge Pack v0.1 applied to Qwen3-14B (Apache-2.0, Alibaba Cloud). It writes and reviews code inside a workgroup framing where groups are isolated security domains and cross-group or cross-site work requires an explicit policy grant, and it grounds THOX platform answers in the Knowledge Pack. The weights are byte-identical to the upstream Qwen3-14B release; THOX work is the configuration profiles, the prompt pack, packaging and documentation. This is not a THOX-trained model.
55
 
@@ -60,6 +101,15 @@ Code generation, review and refactoring assistance for workgroup teams; platform
60
  No training. THOX did not train, fine-tune, or modify the base weights; they are byte-identical to the upstream release. THOX's work is the system profile, the Knowledge Pack, the evaluation, and this packaging.
61
 
62
  ## Evaluation
 
 
 
 
 
 
 
 
 
63
  Measured on 2026-11-01 (x86_64 workstation, 16 GB-class GPU; shared-host execution). Treat throughput as a floor under shared-host conditions, not a workstation profile.
64
  - Base pinned: qwen3:14b, digest sha256:bdbd181c33f2ed1b31c972991882db3cf4d192569092138a7d29e973cd9debe8, 9.28 GB.
65
  - Load: success at 8,192 context. Runner memory footprint while loaded: 10.49 GB (VRAM + system RAM as scheduled by the runner).
@@ -77,3 +127,14 @@ Apache-2.0. The base model is Apache-2.0 licensed; this configuration and packag
77
 
78
  ## Provenance
79
  Upstream: qwen3:14b (digest sha256:bdbd181c33f2ed1b31c972991882db3cf4d192569092138a7d29e973cd9debe8). Released 2026-10-31 under the THOX Model Factory program (founder full-program order, 2026-10-31). Card standard: THOX Model Card Standard v1.0.
 
 
 
 
 
 
 
 
 
 
 
 
18
  - ollama
19
  - configuration-derived
20
  ---
21
+ <div align="center">
22
 
23
+ <img src="hero-family.png" alt="The THOX four-device family" width="100%">
24
 
25
+ <h1>ThoxWork-Code-14B</h1>
26
+
27
+ **THOX Work coding build (Qwen3-14B) — THOX Work build: the coding system profile, the enterprise workgroup prompt pack and the THOX Knowledge Pack v0.1 applied to Qwen3-14B (Apache-2.0, Alibaba Cloud)**
28
+
29
+ **26.28 tok/s long-form (measured 2026-11-01) · 10.49 GB footprint · Grounding 8/8 · Live on Ollama**
30
+
31
+ </div>
32
+
33
+ | ![License](https://img.shields.io/badge/License-Apache_2.0-05A451) | ![Params](https://img.shields.io/badge/Params-14B-05A451) | ![Base](https://img.shields.io/badge/Base-Qwen3--14B-05A451) | ![Access](https://img.shields.io/badge/Access-Public-05A451) |
34
+ |---|---|---|---|
35
+
36
+ **Run it:** [Ollama — Thox-ai/thoxwork-code-14b`).](https://ollama.com/Thox-ai/thoxwork-code-14b`).) · **Find your model:** [THOX Model Factory](https://www.thox.ai/models) · **Collection:** [Code Models](https://huggingface.co/collections/Thox-ai/thox-code-models-6ae7518c544bc6b0fd740cd4)
37
+
38
+ <div align="center">
39
+ <a href="https://www.kickstarter.com/projects/thox/four-devices-one-private-ai-ecosystem-thoxai?ref=4tzhm0"><img src="https://huggingface.co/spaces/Thox-ai/README/resolve/main/back-thox-on-kickstarter.jpg" alt="Back THOX on Kickstarter" width="520"></a>
40
+ </div>
41
 
42
  > **Your AI. Your Data. Your Rules.™**
43
 
44
+ > **Status: PACKAGING ONLY — RUNNABLE BUILD ON OLLAMA.** This repository holds the THOX packaging (Modelfile, manifest, Knowledge Pack) and this card; the weights are the upstream base model's, pulled by Ollama from the pinned base named in the card. THOX did not train these weights.
45
+
46
+ ThoxWork-Code-14B is a THOX-configured model: upstream Qwen3-14B weights with a THOX system profile and the versioned THOX Knowledge Pack v0.1. This repo holds the packaging; the runnable model lives on Ollama (`ollama run Thox-ai/thoxwork-code-14b`). THOX did not train these weights.
47
+
48
+
49
+ ## Highlights
50
+
51
+ - THOX Work build: the coding system profile, the enterprise workgroup prompt pack and the THOX Knowledge Pack v0.1 applied to Qwen3-14B (Apache-2.0, Alibaba Cloud).
52
+ - ```bash ollama run Thox-ai/thoxwork-code-14b ``` Build it yourself from this repo: ```bash hf download Thox-ai/ThoxWork-Code-14B --local-dir ./ThoxWork-Code-14B cd ./ThoxWork-Code-14B && ollama create thoxwork-code-14b -f Modelfile && ollama run thoxwork-code-14b ```
53
+
54
+ ## Specifications
55
+
56
+ | | |
57
+ |---|---|
58
+ | Base model | Qwen/Qwen3-14B |
59
+ | Parameters | 14B |
60
+ | Formats | Ollama Modelfile |
61
+ | License | Apache-2.0. The base model is Apache-2.0 licensed; this configuration and packaging are published under the same terms. |
62
+ | This repository contains | Packaging and documentation — see Files table |
63
 
64
  ## Files
65
 
 
84
  cd ./ThoxWork-Code-14B && ollama create thoxwork-code-14b -f Modelfile && ollama run thoxwork-code-14b
85
  ```
86
 
87
+
88
+ ## Platforms
89
+
90
+ | Platform / device | Run path |
91
+ |---|---|
92
+ | Ollama (laptop / desktop) | `ollama run Thox-ai/thoxwork-code-14b`).` |
93
+ | Hugging Face | Download the builds listed under Files from this repo (see Run it) |
94
  ## Summary
95
  THOX Work build: the coding system profile, the enterprise workgroup prompt pack and the THOX Knowledge Pack v0.1 applied to Qwen3-14B (Apache-2.0, Alibaba Cloud). It writes and reviews code inside a workgroup framing where groups are isolated security domains and cross-group or cross-site work requires an explicit policy grant, and it grounds THOX platform answers in the Knowledge Pack. The weights are byte-identical to the upstream Qwen3-14B release; THOX work is the configuration profiles, the prompt pack, packaging and documentation. This is not a THOX-trained model.
96
 
 
101
  No training. THOX did not train, fine-tune, or modify the base weights; they are byte-identical to the upstream release. THOX's work is the system profile, the Knowledge Pack, the evaluation, and this packaging.
102
 
103
  ## Evaluation
104
+
105
+ Measured by THOX on 2026-11-01 (x86_64 workstation, 16 GB-class GPU, shared-host execution; treat throughput as a floor). Full narrative under Evaluation below.
106
+
107
+ | Benchmark | Result |
108
+ |---|---:|
109
+ | Throughput (long-form, 2,207 tokens) | 26.28 tok/s |
110
+ | Runner footprint (8,192 context) | 10.49 GB |
111
+ | Factory smoke questions | 3/3 |
112
+ | Knowledge Pack grounding | 8/8 topics |
113
  Measured on 2026-11-01 (x86_64 workstation, 16 GB-class GPU; shared-host execution). Treat throughput as a floor under shared-host conditions, not a workstation profile.
114
  - Base pinned: qwen3:14b, digest sha256:bdbd181c33f2ed1b31c972991882db3cf4d192569092138a7d29e973cd9debe8, 9.28 GB.
115
  - Load: success at 8,192 context. Runner memory footprint while loaded: 10.49 GB (VRAM + system RAM as scheduled by the runner).
 
127
 
128
  ## Provenance
129
  Upstream: qwen3:14b (digest sha256:bdbd181c33f2ed1b31c972991882db3cf4d192569092138a7d29e973cd9debe8). Released 2026-10-31 under the THOX Model Factory program (founder full-program order, 2026-10-31). Card standard: THOX Model Card Standard v1.0.
130
+
131
+ ## Citation
132
+
133
+ ```bibtex
134
+ @misc{thoxai2026thoxworkcode14b,
135
+ title={ThoxWork-Code-14B},
136
+ author={THOX.ai},
137
+ year={2026},
138
+ url={https://huggingface.co/Thox-ai/ThoxWork-Code-14B}
139
+ }
140
+ ```