Hacker Newsnew | past | comments | ask | show | jobs | submit | varyherb's commentslogin

Does anyone know of similar solutions that can be self-hosted? (without a 12 service stack like Daytona [1])

[1] https://www.daytona.io/docs/en/oss-deployment/


Have you tried https://orbstack.dev/?


Project Euler spurred my love for math and programming when I was 15. I got through the first few with pencil and paper and then quickly realized I was going to need to learn how to program. Got started with Java (in Eclipse) and then moved on to Python. I feel like it completely changed the trajectory of my life.


I believe this can be configured in Claude Code via the following environment variable:

ANTHROPIC_BETAS="context-1m-2025-08-07" claude


Tested this morning. Worked wonderfully, except ran into output issues. Attempted to patched the minified Claude file's CLAUDE_CODE_MAX_OUTPUT_TOKENS hard limit of 32000 on Sonnet to 64000, which worked, and I was able to generate outputs above 32000 tokens, but this coincided with a breakage of the 1m context window for me. Still testing and playing around with this, but this may be getting patched?


Have you tested it? I see that this env var isn't specified in their docs

https://docs.anthropic.com/en/docs/claude-code/settings#envi...


Yup! Claude Code has a lot of undocumented configuration. Once I saw the beta header value in their docs [1], I tried to see in their source code if there was anyway to specify this flag via env var config. Their source code is already on your computer, just gotta dig through the minified JS :) Try:

`cat $(which claude) | grep ANTHROPIC_BETAS`

Sibling comment's approach with the other (documented) env var works too.

[1] https://docs.anthropic.com/en/docs/build-with-claude/context...


    Add these settings to your `.claude/settings.json`:

     ```json
     {
       "env": {
         "ANTHROPIC_CUSTOM_HEADERS": {"anthropic-beta":
     "context-1m-2025-08-07"},
         "ANTHROPIC_MODEL": "claude-sonnet-4-20250514",
         "CLAUDE_CODE_MAX_OUTPUT_TOKENS": 8192
       }
     }
     ```


I was running into this too until I started upgrading with

llm install -U llm

instead of

uv tool upgrade llm

(the latter of which is recommended by simonw in the original post)


Thanks! I didn't realize "llm install -U llm" did that. I'll add that to the upgrade docs.


Same

  Just shy of a good landing
  Score: 78.3 point landing
  Speed: 5.5mph
  Angle: 0.3°
  Time: 577 seconds
  Flips: 206
  Max speed: 302.4mph
  Max height: 12012ft
  Engine used: 69 times
  Boosters used: 89 times


Perhaps some internet sarcasm is going over my head, but the lyrics are by Samuel Francis Smith. Its melody is the same as that of "God Save the Queen/King."


Where did you find the valuation figure? Or do you just mean the size of the new investment?


It's 80m on 800m valuation. It was on some tweet.


This is a gross oversimplification of the author's detailed (if a bit editorialized) analysis. The author actually acknowledges, and later expounds upon, the apparent incongruity you mentioned:

> I look at communities like Detroit and I understand why the youth there resort to crime, but what the hell is the matter with the Bay Area? Arguably the wealthiest metropolitan area in the United States. A place with too many jobs and not enough people to fill them.


Not to mention 1, 2, and 6 :)

7! = 5040 shows up as well in the OEIS sequence. Buuut, 8! = 40320 does not...too many useless 2's in the prime factorization?


Are you at all concerned about the selection of the workspace ID as the partition key? With every workspace on a single partition, couldn't a high-throughput workspace create a "hot" partition that will negatively impact other workspaces on that partition?

At the cost of potentially introducing more cross-partition queries, you might benefit from splitting up high-throughput workspaces. See strategy in https://d0.awsstatic.com/whitepapers/Multi_Tenant_SaaS_Stora..., pages 17-20.


Perhaps this is why they have 15 logical shards per physical machine. Seems like if one logical shard gets too hot, it could be moved to a new physical machine.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: