I love the Chinese models.

I use DeepSeek exclusively and now Kimi K3 offers a great planning assistant for more advanced coding tasks.

DeepSeek v4 Flash is extremely fast and is able to handle pretty much anything I've thrown at it (I use mostly Rust, PSQL, Angular and Terraform).

I self host Bifrost as my LLM gateway, though I wish LLM vendors would do monthly/daily automatic billing (like VPS providers do) rather than prepaid + auto-top up.

It's annoying maintaining a non-refundable minimum balance across vendors, I would rather be billed for my exact usage.

OpenRouter helps, but I don't really like it as a service and not a fan of the mark up.

▲

arikrahman 2 hours ago | parent | next [-]

The best part is using harnesses like reasonix or whale make cache hit at a rate close to 98%, making requests converge to practically free. And that's with unsubsidized American providers like cloudflare or Digital Ocean.

▲

KronisLV an hour ago | parent | next [-]

In long form tasks, across multiple harnesses (Claude Code, OpenCode, Kimi Code, ZCode) my cache rates are typically 96-99%.

I don’t think that’s particularly out of the ordinary. Do people have different experiences with other harnesses? Which ones?

▲

mlazos an hour ago | parent | prev | next [-]

How can you be hitting cache on what I think are novel LLM prompts …

	▲	oceansweep an hour ago \| parent \| next [-]
		Not them but my understanding is that the harness will send a simple 'heartbeat' message to keep the cache 'warm', (see prefix caching: https://handbook.modular.com/inference-optimization/prefix-c... ) which can then be edited/changed, which does cause the user to incur a fee, but its much less than the amount they'd pay on a no-cache hit request.
	▲	verdverm an hour ago \| parent \| prev [-]
		multi turn sessions, they are typically in the high 90% hit rate across all providers without doing much of anything

▲

sghiassy an hour ago | parent | prev [-]

I’ve never used either of those tools. Do they work well?

▲

skeledrew an hour ago | parent | prev | next [-]

Prepaid means someone can't sign up, use a bunch of inference, then cancel their card and disappear into the sunset. VPS is a more long term investment where it's harder to switch and it doesn't cost the provider much if a few users jump out without paying for a month.

	▲	apatheticonion 26 minutes ago \| parent [-]
		I do understand the reasoning for it, it still sucks as a user who is doing the right thing.

▲

copperx 2 hours ago | parent | prev | next [-]

What don't you like about the service, besides the mark up?

▲

apatheticonion an hour ago | parent | next [-]

For me, the only utility OpenRouter gives me is billing consolidation - I don't really need the routing capabilities because I use Bifrost for that.

As a router, it's not very feature rich. For example I restricted the available models to the ones I want to use however the `/models` endpoint still lists all the models, making my LLM client list the 200+ models available on the service (even though they will throw an error if I try to use them).

With Bifrost, I can also create model aliases with custom configuration - for example I can create a model alias `deepseek-v4-flash-nothink` which disables thinking. I can create `deepseek-v4-flash-caveman` which injects the caveman skill (to save tokens) etc.

Plus I can contribute to Bifrost, which I can't do with OpenRouter.

▲

robbiet480 an hour ago | parent [-]

Did you look at LiteLLM at all? It seems fine but Bifrost looks interesting too.

	▲	apatheticonion an hour ago \| parent [-]
		Yeah I looked at it, LiteLLM is functionally more mature. The only reason I didn't go for it is I'm not a fan of Python dependency management, Bifrost is just a single executable that uses nearly no memory and is lightning fast.

▲

schneehertz 30 minutes ago | parent | prev [-]

openrouter’s API has removed parameters specific to certain models, impairing model functionality.

For example, the image generation parameters of GPT-image-2 are largely ineffective.

▲

ronsor 2 hours ago | parent | prev | next [-]

Have you looked into OpenCode Zen or OpenCode Go?

https://opencode.ai/zen

▲

samtp 2 hours ago | parent [-]

there is also https://cline.bot/cline-pass

	▲	verdverm an hour ago \| parent [-]
		basically the same deal as OC Go, lending credence to token commodification

▲

bellowsgulch 42 minutes ago | parent | prev [-]

> OpenRouter helps, but I don't really like it as a service and not a fan of the mark up.

You're right, I hate it when a business exists to make money. They should just provide their service for free. After all, anyone can build an OpenAI API compatible server that works with every major provider, and even build it using an LLM at that!