Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Changes in the system prompt between Claude Opus 4.6 and 4.7

simonwillison.net

Post reply on HN
Print view View original

131–140 of 240 posts

Re: Changes in the system prompt between Claude Opus 4.6 and 4.7

#131

Post by j-bos »

ikari_pl wrote: post #85

> Claude keeps its responses focused and concise so as to avoid potentially overwhelming the user with overly-long responses. Even if an answer has disclaimers or caveats, Claude discloses them briefly and keeps the majority of its response focused on its main answer. I am strongly opinionated against this. I use Claude in some low-level projects where these answers are saving me from making really silly things, as w…

Agreed. Sprawling system prompts like that are building for the least common denominator, nerfing for anyone or anytime going further.

Re: Changes in the system prompt between Claude Opus 4.6 and 4.7

#132

Post by aesthesia »

sigmoid10 wrote: post #14
mysterydip wrote: post #10

Earlier quoted context omitted.

I assume the reason it’s not baked in is so they can “hotfix” it after release. but surely that many things don’t need updates afterwards. there’s novels that are shorter.

Yeah that was the original idea of system prompts. Change global behaviour without retraining and with higher authority than users. But this has slowly turned into a complete mess, at least for Anthropic. I'd love to see OpenAI's and Google's system prompts for comparison though. Would be interesting to know if they are just more compute rich or more efficient.

Leaked/extracted system prompts for other chat models, particularly ChatGPT, are often around this size. Here's GPT-5.4: https://github.com/asgeirtj/system_prompts_leaks/blob/main/O...

Re: Changes in the system prompt between Claude Opus 4.6 and 4.7

#133

Post by stingraycharles »

j-bos wrote: post #131
ikari_pl wrote: post #85

> Claude keeps its responses focused and concise so as to avoid potentially overwhelming the user with overly-long responses. Even if an answer has disclaimers or caveats, Claude discloses them briefly and keeps the majority of its response focused on its main answer. I am strongly opinionated against this. I use Claude in some low-level projects where these answers are saving me from making really silly things, as w…

Agreed. Sprawling system prompts like that are building for the least common denominator, nerfing for anyone or anytime going further.

You do realize that similar biases are also present in the training data?

Re: Changes in the system prompt between Claude Opus 4.6 and 4.7

#134

Post by stingraycharles »

dhedlund wrote: post #50
chatmasta wrote: post #47

Earlier quoted context omitted.

Claude Code system prompt diffs are available here: https://cchistory.mariozechner.at/?from=2.1.98&to=2.1.112 (URL is to diff since 2.1.98 which seems to be the version that preceded the first reference to Opus 4.7)

The "Picking delaySeconds" section is quite enlightening. I feel like this explains about a quarter to half of my token burn. It was never really clear to me whether tool calls in an agent session would keep the context hot or whether I would have to pay the entire context loading penalty after each call; from my perspective it's one request. I have Claude routinely do large numbers of sequential tool calls, or have…

This is somewhat obvious if you realize that HTTP is a stateless protocol and Anthropic also needs to re-load the entire context every time a new request arrives.

The part that does get cached - attention KVs - is significantly cheaper.

If you read documentation on this, they (and all other LLM providers) make this fairly clear.

Re: Changes in the system prompt between Claude Opus 4.6 and 4.7

#135

Post by pests »

sigmoid10 wrote: post #9

I knew these system prompts were getting big, but holy fuck. More than 60,000 words. With the 3/4 words per token rule of thumb, that's ~80k tokens. Even with 1M context window, that is approaching 10% and you haven't even had any user input yet. And it gets churned by every single request they receive. No wonder their infra costs keep ballooning. And most of it seems to be stable between claude version iterations to…

> And it gets churned by every single request they receive.

Not true, it gets calculated once and essentially baked into initial state basically and gets stored in a standard K/V prefix cache. Processing only happens on new input (minus attention which will have to content with tokens from the prompt)

Re: Changes in the system prompt between Claude Opus 4.6 and 4.7

#136

Post by pests »

sigmoid10 wrote: post #21
formerly_proven wrote: post #13

Earlier quoted context omitted.

Surely the system prompt is cached across accounts?

You can cache K and V matrices, but for such huge matrices you'll still pay a ton of compute to calculate attention in the end even if the user just adds a five word question.

The state of the system can be cached after the system prompt is calculated and all new chats start from that state. O(n^2) is not great but apparently its fine at these context lengths and I'm sure this is a factor in their minimum prompt cost. Advances like grouped query or multi head attention or sparse attention will eventually get rid of that exponential, hopefully.

Re: Changes in the system prompt between Claude Opus 4.6 and 4.7

#137

Post by SilverElfin »

MoltenMan wrote: post #88
zozbot234 wrote: post #79

Earlier quoted context omitted.

That part of the system prompt is just stating that telling someone who has an actual eating disorder to start counting calories or micro-manage their eating in other ways (a suggestion that the model might well give to an average person for the sake of clear argument, which would then be understood sensibly and taken with a grain of salt) is likely to make them worse off, not better off. This seems like a common-sen…

The problem is that this is an incredibly niche / small issue (i.e. At some point you just have to accept that llm's, like people, make mistakes, and that's ok!

Worse, it reveals the kind of moralistic control Anthropic will impose on the world. If they get enough power, manipulation and refusal is the reality everyone will face whenever they veer outside of its built in worldview.

Re: Changes in the system prompt between Claude Opus 4.6 and 4.7

#138

Post by phainopepla2 »

alwillis wrote: post #111
MoltenMan wrote: post #88

Earlier quoted context omitted.

The problem is that this is an incredibly niche / small issue (i.e. At some point you just have to accept that llm's, like people, make mistakes, and that's ok!

>The problem is that this is an incredibly niche / small issue (i.e. It's not a niche issue at all. 29 million people in the US are struggling with an eating disorder [1]. > This single paragraph is going to legitimately cost anthropic at least 4, maybe 5 digits. It's 59 out of 3,791 words total in the system prompt. That's 1.48%. Relax. It should go without saying, but Anthropic has the usage data; they must be seei…

Your source says "Right now, nearly 29 million Americans are struggling with an eating disorder," and then in the table below says that the number of "Americans affected in their lifetime" is 29 million. Two very different things, barely a paragraph apart.

I don't mean to dispute your assertion that it's not a niche issue, but that site does not strike me as a reliable interpreter of the facts.

Re: Changes in the system prompt between Claude Opus 4.6 and 4.7

#139

Post by unshavedyak »

gck1 wrote: post #56
embedding-shape wrote: post #6

> The new section includes: When a request leaves minor details unspecified, the person typically wants Claude to make a reasonable attempt now, not to be interviewed first. Uff, I've tried stuff like these in my prompts, and the results are never good, I much prefer the agent to prompt me upfront to resolve that before it "attempts" whatever it wants, kind of surprised to see that they added that

I even have a specific, non-negotiable phase in the process where model MUST interview me, and create an interview file with everything captured. Plan file it produces must always include this file as an artifact and interview takes the highest precedence. Otherwise, the intent gets lost somewhere in the chat transcript.

This is interesting, can you link any more details on it?

Re: Changes in the system prompt between Claude Opus 4.6 and 4.7

#140

Post by ares623 »

sigmoid10 wrote: post #9

I knew these system prompts were getting big, but holy fuck. More than 60,000 words. With the 3/4 words per token rule of thumb, that's ~80k tokens. Even with 1M context window, that is approaching 10% and you haven't even had any user input yet. And it gets churned by every single request they receive. No wonder their infra costs keep ballooning. And most of it seems to be stable between claude version iterations to…

Does Claude Code (or whatever harness) have it's own system prompt of on top of Opus'?
Post reply on HN