Show HN: I taught GPT-OSS-120B to see using Google Lens and OpenCV

I built an MCP server that gives any local LLM real Google search and now vision capabilities - no API keys needed.

  The latest feature: google_lens_detect uses OpenCV to find objects in an image, crops each one, and sends them to Google Lens for identification. GPT-OSS-120B, a text-only model with
   zero vision support, correctly identified an NVIDIA DGX Spark and a SanDisk USB drive from a desk photo.

  Also includes Google Search, News, Shopping, Scholar, Maps, Finance, Weather, Flights, Hotels, Translate, Images, Trends, and more. 17 tools total.

  Two commands: pip install noapi-google-search-mcp && playwright install chromium

  GitHub: https://github.com/VincentKaufmann/noapi-google-search-mcp
  PyPI: https://pypi.org/project/noapi-google-search-mcp/

Booyah!

21 points | by vkaufmann 3 hours ago

4 comments

magic_hamster 27 minutes ago
> GPT-OSS-120B, a text-only model with zero vision support, correctly identified an NVIDIA DGX Spark and a SanDisk USB drive from a desk photo.
But wasn't it Google Lens that actually identified them?
N_Lens 2 hours ago
Looks like a TOS violation to me to scrape google directly like that. While the concept of giving a text only model 'pseudo vision' is clever, I think the solution in its current form is a bit fragile. The SerpAPI, Google Custom Search API, etc. exist for a reason; For anything beyond personal tinkering, this is a liability.
[-]
- embedding-shape 1 hour ago
  > Looks like a TOS violation to me to scrape google directly like that
  If something was built by violating TOS' and you use that to do more TOS violations against the ones who initially did the TOS violations to build the thing, do they cancel out each other?
  Not about GPT-OSS specifically, but say you used Gemma for the same purpose instead for this hypothetical.
- speedgoose 2 hours ago
  Isn’t SerpAPI about scraping Google through residential proxies as a service ?
  [-]
  - peddling-brink 2 hours ago
    And they are getting sued. https://blog.google/innovation-and-ai/technology/safety-secu...
    [-]
    - interloxia 28 minutes ago
      105 comments from a few months ago.
      https://news.ycombinator.com/item?id=46329109
    - mt42or 1 hour ago
      All ia models are built on stolen data, that's fair war.
tanduv 1 hour ago
you eventually get hit with captcha with the playwright approach
TZubiri 2 hours ago
have you tried Llama? In my experience it has been strictly better than GPT OSS, but it might depend on specifically how it is used.
[-]
- embedding-shape 1 hour ago
  Have you tried GPT-OSS-120b MXFP4 with reasoning effort set to high? Out of all models I can run within 96GB, it seems to consistently give better results. What exact llama model (+ quant I suppose) is it that you've had better results against, and what did you compare it against, the 120b or 20b variant?
  [-]
  - magic_hamster 25 minutes ago
    How are you running this? I've had issues with Opencode formulating bad messages when the model runs on llama.cpp. Jinja threw a bunch of errors and GPT-OSS couldn't make tool calls. There's an issue for this on Opencode's repo but seems like it's been waiting or a couple of weeks.
    > What exact llama model (+ quant I suppose) is it that you've had better results against
    Not llama, but Qwen3-coder-next is on top of my list right now. Q8_K_XL. It's incredible (not just for coding).