Math & Reasoning

“Ranks #1 of 17 on LMArena's overall text arena (Elo 1512), based on blind human preference votes.”

“> It's a pretty high moat getting into stuff like simulation software I'm currently working on a simulation game about space and orbital mechanics. I have a lot of software experience, I know how to build large projects and architect my code, and I know how to to test the end result to ensure I'm getting what I want. But I also don't have a strong math or physics background. In my experience, Claude (Opus 4.6+) has had no issues writing any simulation or game related math code. And the key thin…”

“The initial motivation for this was likely to thwart any competition. Already Anthropic has accused some companies of organized distillation efforts at a massive scale. Back when I used antigravity, it used to show the reasoning intact - at least for Gemini Pro 3.1, and likely for Claude Opus 4.6 (not 100% certain about it). I have some recollection of stopping the models mid-turn when they started going astray. As a power user, I find reasoning fascinating to read and genuinely useful at times…”

“> It's a pretty high moat getting into stuff like simulation software I'm currently working on a simulation game about space and orbital mechanics. I have a lot of software experience, I know how to build large projects and architect my code, and I know how to to test the end result to ensure I'm getting what I want. But I also don't have a strong math or physics background. In my experience, Claude (Opus 4.6+) has had no issues writing any simulation or game related math code. And the key thin…”

“Ranks #4 of 17 on LMArena's overall text arena (Elo 1499), based on blind human preference votes.”

“I was talking to a Claude Opus 4.7 chatbot about low-discrepancy sequences and made the mistake of framing a thought experiment as "deceiving" a vendor who owns a scrambler by making a hard to scramble low-discrepancy sequence generator. Admittedly that sounds sketchy but the conversation up to that point--without the explicit framing--was about the same subject just shorn of metaphor. With the framing I got a refusal message that likened my request to Dual_EC_DRBG!”

“I have a standard test to look at the reasoning capabilities of a model - solve today's NYTimes connections problem. Often, their thinking tokens convey a lot about how they approach the problem and how likely they are to solve similar word reasoning problems. Claude 4.7 and Gemini 3.1 Pro have nailed all so far, GPT 5.5 failed miserably. Of the chinese models, Kimi-K-2.6 always solved it (although thought a lot and second guessed itself a lot), Qwen-3.6-Plus often gave wrong answers and GLM-5.…”

“The initial motivation for this was likely to thwart any competition. Already Anthropic has accused some companies of organized distillation efforts at a massive scale. Back when I used antigravity, it used to show the reasoning intact - at least for Gemini Pro 3.1, and likely for Claude Opus 4.6 (not 100% certain about it). I have some recollection of stopping the models mid-turn when they started going astray. As a power user, I find reasoning fascinating to read and genuinely useful at times…”

“Ranks #9 of 17 on LMArena's overall text arena (Elo 1489), based on blind human preference votes.”

“Ranks #9 of 17 on LMArena's overall text arena (Elo 1489), based on blind human preference votes.”

“Two key quotes: • Reasoning: Through the expansion of reasoning tokens, DeepSeek-V4-Pro-Max demonstrates superior performance relative to GPT-5.2 and Gemini-3.0-Pro on standard reasoning benchmarks. Nevertheless, its performance falls marginally short of GPT-5.4 and Gemini3.1-Pro, suggesting a developmental trajectory that trails state-of-the-art frontier models by approximately 3 to 6 months. Furthermore, DeepSeek-V4-Flash-Max achieves comparable performance to GPT-5.2 and Gemini-3.0-Pro, esta…”

“Ranks #3 of 17 on LMArena's overall text arena (Elo 1499), based on blind human preference votes.”

“Ranks #3 of 17 on LMArena's overall text arena (Elo 1499), based on blind human preference votes.”

“Two key quotes: • Reasoning: Through the expansion of reasoning tokens, DeepSeek-V4-Pro-Max demonstrates superior performance relative to GPT-5.2 and Gemini-3.0-Pro on standard reasoning benchmarks. Nevertheless, its performance falls marginally short of GPT-5.4 and Gemini3.1-Pro, suggesting a developmental trajectory that trails state-of-the-art frontier models by approximately 3 to 6 months. Furthermore, DeepSeek-V4-Flash-Max achieves comparable performance to GPT-5.2 and Gemini-3.0-Pro, esta…”

“"But LLMs are prone to hallucinations which can really impact a string of interdependent logic like a proof. So I’m assuming it would respond with something that’s not complete nonsense to this proof most of the time." Unfortunately in my experience that's not really the case. For me, very often GPT 5.5 (which was a good deal better than Opus at this kind of task) would just get stuck for long periods when working in a logic like Iris. It wouldn't necessarily outright prove nonsense, but it wou…”

“this explains so much why gpt 5.5 has been so bad lately it was really puzzling why it struggled so much where when it first came out it was one shotting stuff totally amazing, i tried the prompt that will tell you if your plan is degraded: codex exec --json --skip-git-repo-check --ephemeral -s read-only --disable memories -m gpt-5.5 -c model_reasoning_effort=high "Do not use external tools. A black bag contains candies with counts: round apple 7, round peach 9, round watermelon 8; star apple 7…”

“Ranks #6 of 17 on LMArena's overall text arena (Elo 1491), based on blind human preference votes.”

“I mean you can just ask them to do exactly that. Especially with GPT (5.5), I've been having a lot of issues with it just repeatedly stalling out. I had to build a quota monitoring skill so that it'd keep plowing forward until either the task was finished (in some way) or the quota budget was exhausted. I also had issues with the compaction. Codex seems to compact... weirdly, resulting in the agent becoming a newborn after each compaction event. Telling it to use a notes file is basically essen…”

“"But LLMs are prone to hallucinations which can really impact a string of interdependent logic like a proof. So I’m assuming it would respond with something that’s not complete nonsense to this proof most of the time." Unfortunately in my experience that's not really the case. For me, very often GPT 5.5 (which was a good deal better than Opus at this kind of task) would just get stuck for long periods when working in a logic like Iris. It wouldn't necessarily outright prove nonsense, but it wou…”

“this explains so much why gpt 5.5 has been so bad lately it was really puzzling why it struggled so much where when it first came out it was one shotting stuff totally amazing, i tried the prompt that will tell you if your plan is degraded: codex exec --json --skip-git-repo-check --ephemeral -s read-only --disable memories -m gpt-5.5 -c model_reasoning_effort=high "Do not use external tools. A black bag contains candies with counts: round apple 7, round peach 9, round watermelon 8; star apple 7…”