Hugging Face's logo

Spaces:
turing-softworks
/
README
Running

App Files Community
README
File size: 2,783 Bytes
20bc994
 
 
 
 
 
346135c
20bc994
 
346135c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
---
title: README
emoji: 📊
colorFrom: indigo
colorTo: indigo
sdk: static
pinned: true
---

# 🌱 turing.garden

**AI for the masses.**

`turing.garden` is an AI research and engineering company focused on creating accessible, efficient, and practical language models.

We believe that in today and tomorrow's landscape, LLMs are becoming increasingly:

* expensive (both to train and to compute);
* power hungry;
* and inaccessible to ordinary people.

---

## Our mission

We aim to make advanced AI available to everyone by creating models that are:

* affordable to deploy and use;
* optimised for accuracy and speed;
* more energy-efficient;
* suitable for consumer hardware;
* and built with safety in mind.

---

## What we build

### Foundation models

General-purpose LLMs and SLMs trained for reasoning, understanding, and knowledge

### Specialist models

Instead of forcing one model to excel at every domain, we develop specialist models for specific domains, such as:

* software engineering;
* biology, physics, chemistry;
* and general reasoning, etc.

### Distilled Models

Our larger models are distilled into smaller variants that only contain millions of parameters, making them suitable for:

* mobile devices;
* resource-constrained devices;
* and more.

---

## Research

Our work spans research and development, including:

* alignment;
* quantisation;
* efficient training and inference;
* agentic AI;
* and distillation, etc.

---

## Open models

We support open AI where it benefits both researchers and developers.

Most models, datasets, etc, will be released under **OpenRAIL**.

Certain proprietary models and research may remain closed where appropriate.

---

### Targets

We design for:

* consumer-grade hardware;
* workstations;
* and enterprise servers.

We believe AI should not require hyperscale infrastructure whenever possible.

---

## Safety

The brain is one of the most impressive structures in the entire universe.

AI should augment human thinking and should not act as a replacement.

Our research is guided by:

* safety;
* transparency;
* and accessibility.

We do not intend for our models to facilitate:

* warfare;
* autonomous harmful systems;
* the replacement of human critical thinking.

---

## Beyond AI

We also create develop other software!

Check us out at [our website](https://turing.garden).

---

## Philosophy

Bigger AI is not necessarily better AI. Success is measured by impact.

Through research, we believe language models can become more useful, efficient, and accessible for everyone.

---

## Coming Soon

* Foundation language models
* Domain-specific specialist models
* Distilled lightweight models
* Technical reports
* Research publications
* Benchmarks
* Training tools
* Community resources

---