Spaces:

Eklavya16
/

Shortlist

Sleeping

App Files Files Community

Shortlist / backend

Commit History

Aug 30

18:02
Switch OpenRouter default to Nemotron 3 Super for extreme inference speed and strict JSON formatting

Eren-Sama

56c0c14
17:50
Fix OpenRouter fallback chain by using confirmed free models (prevents 404s breaking the chain)

Eren-Sama

84849df
17:36
Implement OpenRouter fallback chain to seamlessly handle 429 rate limit errors from free models

Eren-Sama

322d8cd
17:20
Update default OpenRouter model to Gemma 4 31B (proven free generalist)

Eren-Sama

a139b55
17:13
Update default OpenRouter free model to Llama 3.3 70B (exceptional generalist, 128k context)

Eren-Sama

dac7a70
17:06
Update default OpenRouter free model to gemini-2.0-flash-exp:free (1M context)

Eren-Sama

3bfbd69
17:00
Use gemini-2.0-flash-exp:free on OpenRouter (1M context, best free model)

Eren-Sama

00fd235
16:45
Replace NVIDIA with OpenRouter — stable free models, configurable via env var

Eren-Sama

fd0d0f2
15:52
Switch NVIDIA model to llama-3.1-70b-instruct (fast, 128k, active)

Eren-Sama

a5ac4b2
15:44
Update NVIDIA model to nemotron-3-ultra-550b-a55b (1M context, active)

Eren-Sama

e407887
08:57
Make NVIDIA primary provider, Gemini fallback (Gemini free tier: 20 req/day)

Eren-Sama

42edebc
08:50
Cap repo analysis to 20 files / 4k chars to fit 128k context fallbacks

Eren-Sama

6816543
08:47
Update NVIDIA model to llama-3.3-nemotron-super-49b-v1 (non-deprecated)

Eren-Sama

05dc1d0
08:41
Fix NVIDIA model name to llama-3.1-nemotron-ultra-253b-v1

Eren-Sama

dbc7a0c
08:38
Remove MiniMax M3, keep Gemini + NVIDIA Nemotron fallback chain

Eren-Sama

1e58c96
08:28
Add NVIDIA Nemotron Ultra and MiniMax M3 as 1M-context fallbacks

Eren-Sama

ac80c94
08:14
Replace Groq/OpenAI with Cohere for 128k context fallback

Eren-Sama

57bcb6f
08:03
Configure massive-context Gemini models as fallbacks

Eren-Sama

a3b8cc9
07:51
Fix CI Docker tag case and relax pip dependency constraints

Eren-Sama

2893bc3
07:44
Add missing langchain dependencies to requirements.txt

Eren-Sama

2431d6e
07:33
Fix pylint no-member error for ALLOWED_ORIGINS

Eren-Sama

6ff6053
06:15
Switch LLM provider to Google Gemini 3.7 Flash

Eren-Sama

6f8c1bf

Aug 29

16:58
Change model from 'openai/gpt-oss-120b' to 'llama-3.1-8b-instant'

Eklavya Mohan Agrawal

Unverified signature
1111233
16:58
Update LLM models to llama-3.1-8b-instant

Eklavya Mohan Agrawal

Unverified signature
cf3eeac

Aug 27

21:07
Update .env.example

Eklavya Mohan Agrawal

Unverified signature
65e22bc
21:07
Update portfolio.py

Eklavya Mohan Agrawal

Unverified signature
bb3e988
20:59
Update LLM models to OpenAI GPT-OSS 120B

Eklavya Mohan Agrawal

Unverified signature
659e563

Feb 11

15:00
Update Dockerfile

Eklavya Mohan Agrawal

Unverified signature
1f772d4
14:58
Update Dockerfile

Eklavya Mohan Agrawal

Unverified signature
72cb842
14:54
Update Dockerfile

Eklavya Mohan Agrawal

Unverified signature
aef9899
14:52
Update Dockerfile

Eklavya Mohan Agrawal

Unverified signature
9f65a5e
14:40
Update Dockerfile

Eklavya Mohan Agrawal

Unverified signature
2914f34
14:36
Update Dockerfile

Eklavya Mohan Agrawal

Unverified signature
4b8c8e3
14:31
Update security.py

Eklavya Mohan Agrawal

Unverified signature
3453195
14:30
Update Dockerfile

Eklavya Mohan Agrawal

Unverified signature
4b953e4
14:23
Update Dockerfile

Eklavya Mohan Agrawal

Unverified signature
e0dc19e
14:03
Add root route to main application

Eklavya Mohan Agrawal

Unverified signature
9aa23f2
13:58
Simplify CMD in Dockerfile for uvicorn⁠

Eklavya Mohan Agrawal

Unverified signature
228f5f6
13:54
Modify CORS settings to allow all methods and headers⁠

Eklavya Mohan Agrawal

Unverified signature
9ed2ab9
13:49
Update CMD instruction in Dockerfile for uvicorn

Eklavya Mohan Agrawal

Unverified signature
accc0ea
13:45
Change Uvicorn port to use environment variable PORT

Eklavya Mohan Agrawal

Unverified signature
0ed0afc
13:35
Fix ALLOWED_ORIGINS declaration in config.py

Eklavya Mohan Agrawal

Unverified signature
40dd5bb
13:29
Allow CORS preflight requests in dispatch method⁠

Eklavya Mohan Agrawal

Unverified signature
2089f1d
13:18
Update ALLOWED_ORIGINS for production deployment

Eklavya Mohan Agrawal

Unverified signature
5c0f092
09:24
Initial commit — full-stack AI portfolio architect⁠

Eren-Sama

53e1531

Commit History

Aug 30

Switch OpenRouter default to Nemotron 3 Super for extreme inference speed and strict JSON formatting

Fix OpenRouter fallback chain by using confirmed free models (prevents 404s breaking the chain)

Implement OpenRouter fallback chain to seamlessly handle 429 rate limit errors from free models

Update default OpenRouter model to Gemma 4 31B (proven free generalist)

Update default OpenRouter free model to Llama 3.3 70B (exceptional generalist, 128k context)

Update default OpenRouter free model to gemini-2.0-flash-exp:free (1M context)

Use gemini-2.0-flash-exp:free on OpenRouter (1M context, best free model)

Replace NVIDIA with OpenRouter — stable free models, configurable via env var

Switch NVIDIA model to llama-3.1-70b-instruct (fast, 128k, active)

Update NVIDIA model to nemotron-3-ultra-550b-a55b (1M context, active)

Make NVIDIA primary provider, Gemini fallback (Gemini free tier: 20 req/day)

Cap repo analysis to 20 files / 4k chars to fit 128k context fallbacks

Update NVIDIA model to llama-3.3-nemotron-super-49b-v1 (non-deprecated)

Fix NVIDIA model name to llama-3.1-nemotron-ultra-253b-v1

Remove MiniMax M3, keep Gemini + NVIDIA Nemotron fallback chain

Add NVIDIA Nemotron Ultra and MiniMax M3 as 1M-context fallbacks

Replace Groq/OpenAI with Cohere for 128k context fallback

Configure massive-context Gemini models as fallbacks

Fix CI Docker tag case and relax pip dependency constraints

Add missing langchain dependencies to requirements.txt

Fix pylint no-member error for ALLOWED_ORIGINS

Switch LLM provider to Google Gemini 3.7 Flash

Aug 29

Change model from 'openai/gpt-oss-120b' to 'llama-3.1-8b-instant'

Update LLM models to llama-3.1-8b-instant

Aug 27

Update .env.example

Update portfolio.py

Update LLM models to OpenAI GPT-OSS 120B

Feb 11

Update Dockerfile

Update Dockerfile

Update Dockerfile

Update Dockerfile

Update Dockerfile

Update Dockerfile

Update security.py

Update Dockerfile

Update Dockerfile

Add root route to main application

Simplify CMD in Dockerfile for uvicorn⁠

Modify CORS settings to allow all methods and headers⁠

Update CMD instruction in Dockerfile for uvicorn

Change Uvicorn port to use environment variable PORT

Fix ALLOWED_ORIGINS declaration in config.py

Allow CORS preflight requests in dispatch method⁠

Update ALLOWED_ORIGINS for production deployment

Initial commit — full-stack AI portfolio architect⁠