87 points Edward40 1 hour ago 32 comments

guanlan 1 hour ago | parent

Hi HN, I’m Guanlan, founder of Runta. The eval runs on our runtime, so weigh this accordingly.

The short version: when you change the model, think about changing the harness too. On any one model, cost per pass varies by up to 4.4x

What we ran: 22 harness configurations × 5 open-weight models × 62 terminal tasks × 4 attempts. 27,280 runs, 83B tokens. Each harness ran with default settings at maximum reasoning effort, each trial in a fresh sandbox. Costs are Baseten list prices times reported tokens.

What we found: The lowest-cost harness changes with the model. Codex CLI ranks 1st on cost with GLM-5.3 and 15th with Kimi-K3. Paying more doesn’t buy passes. The harness with the highest pass rate ranked anywhere from 4th to 19th on cost. The model still sets most of the bill (93% of cost variation). The harness is the second lever. Starting prompt size doesn’t predict cost. Output tokens do. 64% of inference spend went to runs that didn’t pass.

radlad 58 minutes ago | parent

Hi Guanlan, I vouched for your comment since you involved in this post, but you should be aware that all of your posts have been flagged, likely for AI patterns in your writing.

> Don't post generated text or AI-edited text. HN is for conversation between humans.

https://news.ycombinator.com/newsguidelines.html

latentsea 38 minutes ago | parent

This is why for open-weights models at least harness aware post-training is becoming a thing. Also closed source harnesses that aren't fully editable are basically an anti-pattern. The labs should just hurry up and productionize self-evolving harnesses, and have the models internalize harness engineering, so they can optimise the harness for the given task. These are solvable problems.

Majromax 42 minutes ago | parent

Saying that 'steps per pass' explains cost seems nearly tautological. Run costs are proportional to the number of tokens produced, and the number of tokens is strongly correlated to the number of steps.

The 'steps per pass' criterion is also not something under the harness's direct control in the way that its tool collection and system prompt (/start tokens) are. Even the cache hit rate is hard to control beyond a limit, since modern harnesses rarely rewrite history.

I'm also suspicious about the run-to-run variance between models. Given the high variability within the presented results and the observation that failed runs (which can spin for a long time before delivering a wrong answer) account for a majority of the cost, I'm not sure that many of the conclusions here would hold up under a second sample.

Zigurd 35 minutes ago | parent

I find myself pulled in two directions by the performance of coding agents: First, keeping them on a short leash seems sensible. I don't get failed results very often. When I first started using coding agents, this seemed not just sensible but required in order to not deal with a lot of hallucinated weirdness. Secondly though, it's pretty cool to vaguely wave your hands in a certain direction and get both user interface and business logic that's objectively nicer than I would've gotten from the effort I would've put into manually coding. Whenever I cautiously use a coding agent, I wonder what I am missing by not letting it cook.

embedding-shape 31 minutes ago | parent

Not every problem should (or even could sometimes) be solved the same way. Are you planning to maintain the produced code for more than a year? No, then wave away. But if you're expecting others to depend on it, and you want to be able to continue evolving what it is, then a tight leash and full understanding of how it actually works, is very much needed, even if you don't actually type characters into an editor anymore.

Otherwise you still end up with balls of spaghetti where each change takes longer and longer to work into the codebase. With a bit of planning, you can keep tackling changes and they keep taking the same amount of time more or less.

smallnamespace 25 minutes ago | parent

The cost of AI is rapidly falling while its engineering capabilities keep rising. This implies that code is becoming more fungible, and you should expect any code you write today to mostly be replaced in the near future.

The requirements will be more stable than the code. Focus on them instead.

catdog 16 minutes ago | parent

> The cost of AI is rapidly falling while its engineering capabilities keep rising.

Wouldn't count on it. There likely is a ceiling, we might or might not be near it.

datsci_est_2015 15 minutes ago | parent

Robust systems experience very little turnover in nearly every non-software discipline that involves systems. Why the hell would we insert risk into our system-building processes by needlessly turning over code?

Zigurd 13 minutes ago | parent

> The requirements will be more stable than the code.

These kinds of changes to the way teams are formed and managed, and how products are developed is what really fascinates me about the impact of coding agents.

Even if it works, for some definition of works, to dump a huge spec into a coding agent context window, I suspect this is where a lot of horror stories about agentic coding come from.

Teams, and the management structure above them, are going to have to get used to more collaborative and iterative product definition and development in order to extract the most value from coding agents.

rrr_oh_man 28 minutes ago | parent

> objectively nicer than I would've gotten from the effort I would've put into manually coding

I find that only works with topics I know very little about.

datsci_est_2015 17 minutes ago | parent

This comment could be interpreted at least two ways, and I’m not sure which it is:

  - tongue in cheek: in topics you know very little about, you don’t have the ability to tell whether the LLM output is trash or not
  - very serious: LLMs allow you to create things you could not have created before due to not having the knowledge
Sort of like a Dunning-Kruger or Gell-Mann amnesia.

devmor 9 minutes ago | parent

I don't see why both can't be true.

LLMs are allowing people to create things with very little effort compared to before, and those creations are of questionable quality - indeterminable by the LLM-user due to a lack of experience.

contravariant 15 minutes ago | parent

I think Conway's law applies. In an ideal situation your code mimics your organisational structure.

The places where you let an agent do its thing is bound by the (implicit) interface your prompt creates. In an ideal situation this interface is a simple isolated thing in your application.

If you just let it do its thing, expect a monolith. You shouldn't rely on it coming up with interfaces that mesh well with what you had in mind, and each of those interfaces represents a decision that it took, restricting future iterations of the code. So a monolith is not only what you can expect, but also the best you can hope for.

If you want to let it cook, create more places for code that does a thing and you don't care how.

Ericson2314 13 minutes ago | parent

BTW the usual point of Conway's structure is to try to adjust the org / who is working on the software, so you don't get a bad software architecture.

The "objective" nature of code is revealing a flaw in the human division of labor.

cronin101 25 minutes ago | parent

I'm far behind the curve with local/open weight models, but I thought it was common practice that you should do some last-mile fine-tuning on the particular harness that you're using and would get a reasonable return on investment for doing so (particularly on quantized models)?

Can anyone confirm/deny this is the case?

If so, it seems a shame not to include this practice into the reporting methodology for full transparency.

jrmg 21 minutes ago | parent

What does "last-mile fine-tuning on [a] harness" mean?

cronin101 18 minutes ago | parent

Taking the open weight model and altering those weights in order to overfit for success on the tool shape / quirks of the harness you are running the model within.

As with most post-training this would be done via some scored/verifiable problem set (we can probably tell if the tool calls blew up during the coding trajectory) e.g. a known set of input tasks where solutions can be verified.

Last-mile aka doing in yourself (on box) vs the compute cluster that created those open weights.

https://en.wikipedia.org/wiki/Fine-tuning_(deep_learning)

I believe (at least this used to be the case) that heavily quantized models that are designed to fit on consumer hardware have inflated failure rates and this can claw back quite a bit of reliability.

ford 15 minutes ago | parent

I don't think this is common, though it would be interesting if ThinkingMachines/HuggingFace/OpenRouter/BaseTen offered fine-tuned variants of the most popular open models that worked with specific harnesses

cronin101 9 minutes ago | parent

In hindsight I think I overestimated how common/achievable this is with the logic that if PewDiePie could do it, so could a layman https://felloai.com/pewdiepies-ai-model-beats-gpt-4o/

However, the hardware requirements are still quite enormous.

I could see a gap in the market for crowdfunded (gofundme?) post-training of open-weight models, per harness. I estimate that the one-time compute cost is probably $x000 so this scales nicely with a large number of interested developers that could reuse the weights.

jrmg 21 minutes ago | parent

"In 27,280 coding agent runs, 64% of inference spend went to runs that failed" (the title this submission was submitted with) is a pretty editorialized title.

The page's title is: "Introducing FrontierHarness Eval v1.1"

ford 18 minutes ago | parent

To be clear - "Failed" here means "Did not meet all pass criteria on a given benchmark task", not "Experienced a runtime failure"

scubazealous 15 minutes ago | parent

While this could be better, I wonder how much time the average developer spends on runs that fail? How much time do students spend on runs that fail?

seizethecheese 11 minutes ago | parent

The headline here is confusing. They ran a bunch of harness and agent combinations on terminal bench tasks, and 64% of the spending was on tasks that didn’t pass terminal bench verifiers.

It’s unclear whether that’s supposed to be surprising against the selected models for the benchmark. I suspect it’s not and the headline is written to pander to “wasting money on AI” upvotes.

andai 8 minutes ago | parent

>Through our experiments, we found that you should consider switching harnesses when switching models to maximize your experience.

I've seen this for years, and seen several papers that studied it: the same prompt will help with one LLM and harm performance for a different LLM on the same task.

And on benchmarks which test LLM x Harness, they find that LLMs tend to do way better with their "own" harness. (Or in the case of Chinese LLMs, the harness of the LLM whose training data they borrowed :)

shevy-java 8 minutes ago | parent

Well, it is a milking system. You have to pay for those tokens anyway.

colinmarc 6 minutes ago | parent

How are people even using Baseten for coding agents? I am using Pi (so very small requests) in a "normal" single-agent way and I hit the rate limits on throughput immediately.