Lie24 commited on
Commit
e6f8d5b
·
verified ·
1 Parent(s): 3e2e456

Rename to StartLux-Decision-v1

Browse files
Files changed (1) hide show
  1. README.md +13 -13
README.md CHANGED
@@ -8,22 +8,22 @@ tags:
8
  - classification
9
  ---
10
 
11
- # StartLux-Decision-2B
12
 
13
  <p align="center"><img src="assets/hero.png" alt="StartLux-Decision: a probability for every option" width="100%"></p>
14
 
15
- StartLux-Decision-2B answers typed questions about a state: pick one of several options, yes or no, or a rating on a scale. The
16
  state can be text, JSON or images, up to 262,144 tokens (256K). Every
17
  question comes back with a probability for each option. Requests and responses use the TypeSafe `/v1/systemone`
18
  format, so clients written for Jev work unchanged.
19
 
20
- Sizes: [0.8B](https://huggingface.co/startlux-models/StartLux-Decision-0.8B) · **2B** · [4B](https://huggingface.co/startlux-models/StartLux-Decision-4B) · [9B](https://huggingface.co/startlux-models/StartLux-Decision-9B) · [27B](https://huggingface.co/startlux-models/StartLux-Decision-27B) · [35B-A3B](https://huggingface.co/startlux-models/StartLux-Decision-35B-A3B)
21
 
22
  ## Results
23
 
24
  <div align="center">
25
 
26
- | StartLux-Decision-2B | |
27
  |:---:|:---:|
28
  | Decision Index 0.2.1 | 44.19 |
29
  | Decision Index 0.2 | 40.72 |
@@ -47,17 +47,17 @@ Latency is end to end over HTTP on one H200 in bf16, one request at a time.
47
 
48
  | Model | JevBench public, of 231 | Intern avg | DI 0.2 / 0.2.1 | Latency, 3 questions |
49
  |:---:|:---:|:---:|:---:|:---:|
50
- | StartLux-Decision-35B-A3B | 210 | 92.29 | 57.24 / 61.55 | 52.5 ms |
51
- | StartLux-Decision-27B | 208 | 91.82 | 59.54 / 63.88 | 102.3 ms |
52
- | StartLux-Decision-9B | 201 | 91.08 | 54.37 / 58.63 | 35.7 ms |
53
- | StartLux-Decision-4B | 204 | 91.17 | 48.38 / 52.75 | 26.0 ms |
54
  | Intern-Decision-4B | 201 | 90.02 | 35.90 / 37.81 | 44.2 ms ¹ |
55
  | JevK5 | 200 | 85.16 | 36.44 / 38.81 | |
56
  | Jev 1.13 | 199 | 88.74 | 51.67 / 57.91 | 64.0 ms ² |
57
- | **StartLux-Decision-2B** | 196 | 88.46 | 40.72 / 44.19 | 15.5 ms |
58
  | SemIf | 187 | 84.23 | 25.70 / 25.94 | |
59
  | Intern-Decision-2B | 180 | 84.68 | 19.49 / 19.38 | 33.3 ms ¹ |
60
- | StartLux-Decision-0.8B | 179 | 85.03 | 35.57 / 38.86 | 12.2 ms |
61
  | Intern-Decision-0.8B | 163 | 79.38 | 11.32 / 11.94 | 34.0 ms ¹ |
62
  | Laya | 130 | 57.77 | 5.51 / 6.04 | |
63
 
@@ -96,8 +96,8 @@ TypeSafe's definitions (for a choice, (p_max − 1/n) / (1 − 1/n)); the top pr
96
  ## Usage
97
 
98
  ```bash
99
- hf download startlux-models/StartLux-Decision-2B --local-dir StartLux-Decision-2B
100
- cd StartLux-Decision-2B
101
  pip install -r requirements.txt
102
  python -m startlux_decision.check . # must print "fast kernels: active"
103
  python -m startlux_decision.server --model . --port 8090
@@ -141,4 +141,4 @@ answers, usage = m.decide("Photo taken at delivery: <image>",
141
  The model weights are released under [CC BY-NC 4.0](https://creativecommons.org/licenses/by-nc/4.0/): free for
142
  research and other non-commercial use, with attribution. Commercial use requires a separate license from
143
  StartLux Labs; contact [contact@startlux.com](mailto:contact@startlux.com). The inference code in
144
- `startlux_decision/` is Apache-2.0. See [LICENSE](https://huggingface.co/startlux-models/StartLux-Decision-2B/blob/main/LICENSE) and [NOTICE](https://huggingface.co/startlux-models/StartLux-Decision-2B/blob/main/NOTICE).
 
8
  - classification
9
  ---
10
 
11
+ # StartLux-Decision-v1-2B
12
 
13
  <p align="center"><img src="assets/hero.png" alt="StartLux-Decision: a probability for every option" width="100%"></p>
14
 
15
+ StartLux-Decision-v1-2B answers typed questions about a state: pick one of several options, yes or no, or a rating on a scale. The
16
  state can be text, JSON or images, up to 262,144 tokens (256K). Every
17
  question comes back with a probability for each option. Requests and responses use the TypeSafe `/v1/systemone`
18
  format, so clients written for Jev work unchanged.
19
 
20
+ Sizes: [0.8B](https://huggingface.co/startlux-models/StartLux-Decision-v1-0.8B) · **2B** · [4B](https://huggingface.co/startlux-models/StartLux-Decision-v1-4B) · [9B](https://huggingface.co/startlux-models/StartLux-Decision-v1-9B) · [27B](https://huggingface.co/startlux-models/StartLux-Decision-v1-27B) · [35B-A3B](https://huggingface.co/startlux-models/StartLux-Decision-v1-35B-A3B)
21
 
22
  ## Results
23
 
24
  <div align="center">
25
 
26
+ | StartLux-Decision-v1-2B | |
27
  |:---:|:---:|
28
  | Decision Index 0.2.1 | 44.19 |
29
  | Decision Index 0.2 | 40.72 |
 
47
 
48
  | Model | JevBench public, of 231 | Intern avg | DI 0.2 / 0.2.1 | Latency, 3 questions |
49
  |:---:|:---:|:---:|:---:|:---:|
50
+ | StartLux-Decision-v1-35B-A3B | 210 | 92.29 | 57.24 / 61.55 | 52.5 ms |
51
+ | StartLux-Decision-v1-27B | 208 | 91.82 | 59.54 / 63.88 | 102.3 ms |
52
+ | StartLux-Decision-v1-9B | 201 | 91.08 | 54.37 / 58.63 | 35.7 ms |
53
+ | StartLux-Decision-v1-4B | 204 | 91.17 | 48.38 / 52.75 | 26.0 ms |
54
  | Intern-Decision-4B | 201 | 90.02 | 35.90 / 37.81 | 44.2 ms ¹ |
55
  | JevK5 | 200 | 85.16 | 36.44 / 38.81 | |
56
  | Jev 1.13 | 199 | 88.74 | 51.67 / 57.91 | 64.0 ms ² |
57
+ | **StartLux-Decision-v1-2B** | 196 | 88.46 | 40.72 / 44.19 | 15.5 ms |
58
  | SemIf | 187 | 84.23 | 25.70 / 25.94 | |
59
  | Intern-Decision-2B | 180 | 84.68 | 19.49 / 19.38 | 33.3 ms ¹ |
60
+ | StartLux-Decision-v1-0.8B | 179 | 85.03 | 35.57 / 38.86 | 12.2 ms |
61
  | Intern-Decision-0.8B | 163 | 79.38 | 11.32 / 11.94 | 34.0 ms ¹ |
62
  | Laya | 130 | 57.77 | 5.51 / 6.04 | |
63
 
 
96
  ## Usage
97
 
98
  ```bash
99
+ hf download startlux-models/StartLux-Decision-v1-2B --local-dir StartLux-Decision-v1-2B
100
+ cd StartLux-Decision-v1-2B
101
  pip install -r requirements.txt
102
  python -m startlux_decision.check . # must print "fast kernels: active"
103
  python -m startlux_decision.server --model . --port 8090
 
141
  The model weights are released under [CC BY-NC 4.0](https://creativecommons.org/licenses/by-nc/4.0/): free for
142
  research and other non-commercial use, with attribution. Commercial use requires a separate license from
143
  StartLux Labs; contact [contact@startlux.com](mailto:contact@startlux.com). The inference code in
144
+ `startlux_decision/` is Apache-2.0. See [LICENSE](https://huggingface.co/startlux-models/StartLux-Decision-v1-2B/blob/main/LICENSE) and [NOTICE](https://huggingface.co/startlux-models/StartLux-Decision-v1-2B/blob/main/NOTICE).