@jutge.org/toolkit 4.4.46 → 4.4.50
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/dist/index.js +579 -580
- package/docs/getting-started-guide.md +2 -2
- package/docs/jutge-ai.md +5 -6
- package/docs/problem-anatomy.md +1 -1
- package/package.json +1 -1
|
@@ -50,7 +50,7 @@ Before we dive into configuration, it's important to know how to get help:
|
|
|
50
50
|
|
|
51
51
|
```bash
|
|
52
52
|
jtk ask "How to create a new problem?"
|
|
53
|
-
jtk ask "Com puc crear un problema nou?" --model "openai/gpt-
|
|
53
|
+
jtk ask "Com puc crear un problema nou?" --model "openai/gpt-6-luna"
|
|
54
54
|
```
|
|
55
55
|
|
|
56
56
|
For this to work, you need to have set up an AI API key (see your platform-specific installation guide).
|
|
@@ -84,7 +84,7 @@ Before using the toolkit, you should configure it with your preferences:
|
|
|
84
84
|
2. **Set a configuration value:**
|
|
85
85
|
|
|
86
86
|
```bash
|
|
87
|
-
jtk config set defaultModel "
|
|
87
|
+
jtk config set defaultModel "openai/gpt-6-luna"
|
|
88
88
|
```
|
|
89
89
|
|
|
90
90
|
3. **Edit configuration interactively:**
|
package/docs/jutge-ai.md
CHANGED
|
@@ -43,8 +43,9 @@ Google Gemini is fast and free for UPC users but its rate limits are so low it i
|
|
|
43
43
|
|
|
44
44
|
Available models:
|
|
45
45
|
|
|
46
|
-
- google/gemini-2.5-flash
|
|
47
46
|
- google/gemini-2.5-flash-lite
|
|
47
|
+
- google/gemini-2.5-flash
|
|
48
|
+
- google/gemini-2.5-pro
|
|
48
49
|
|
|
49
50
|
### OpenAI GPT
|
|
50
51
|
|
|
@@ -52,16 +53,14 @@ OpenAI is slower but more reliable and has a higher rate limit. However, it is n
|
|
|
52
53
|
|
|
53
54
|
Available models:
|
|
54
55
|
|
|
55
|
-
- openai/gpt-
|
|
56
|
-
- openai/gpt-
|
|
57
|
-
- openai/gpt-4.1-nano
|
|
58
|
-
- openai/gpt-4.1-mini
|
|
56
|
+
- openai/gpt-6-luna
|
|
57
|
+
- openai/gpt-6-sol
|
|
59
58
|
|
|
60
59
|
See https://platform.openai.com/docs/pricing for the pricing of the OpenAI models.
|
|
61
60
|
|
|
62
61
|
### Recommendation
|
|
63
62
|
|
|
64
|
-
|
|
63
|
+
Do not use Gemini with UPC key. Try to use `gpt-6-luna` for the quickest (and cheaper!) results. If you need more power, use `gpt-6-sol`.
|
|
65
64
|
|
|
66
65
|
# Jutge<sup>AI</sup> costs
|
|
67
66
|
|
package/docs/problem-anatomy.md
CHANGED
|
@@ -355,7 +355,7 @@ In addition, there must be no spaces before a line break. In particular, there m
|
|
|
355
355
|
|
|
356
356
|
### Sample test cases
|
|
357
357
|
|
|
358
|
-
Sample test cases
|
|
358
|
+
Sample test cases contain `sample` (yes: _contain_, no _start with_) in their filename and will be revealed to users in the problem statement. As such, they should make clear the format of the input and the output for the problem and should be reasonably short.
|
|
359
359
|
|
|
360
360
|
### Public test cases
|
|
361
361
|
|