| | Knowledge vs. wisdom: asking AI "What mushroom is that?" (quesma.com) |
| 2 points by stared 23 hours ago | past | discuss |
|
| | Benchmarking Qwen3.8 27B quantizations: 4-bit holds up, 1-bit collapses (quesma.com) |
| 284 points by stared 1 day ago | past | 136 comments |
|
| | Benchmarking Qwen3.8 27B quantizations: 4-bit holds up, 1-bit collapses (quesma.com) |
| 3 points by stared 6 days ago | past | 1 comment |
|
| | Mushroom hunting with LLMs: what can go wrong? (quesma.com) |
| 56 points by stared 7 days ago | past | 75 comments |
|
| | Benchmarking Qwen3.8 27B quantizations: 4-bit holds up, 1-bit collapses (quesma.com) |
| 3 points by stared 12 days ago | past | 2 comments |
|
| | OpenAI Codex pricing: the $270 PR a $200/month sub covers daily (quesma.com) |
| 3 points by jakozaur 13 days ago | past | discuss |
|
| | Benchmarking Qwen3.8 27B quantizations: 4-bit holds up, 1-bit collapses (quesma.com) |
| 15 points by stared 14 days ago | past | 2 comments |
|
| | Gemini 3.7 Flash, Grok 4.6, GLM-5.3 and DeepSeek V4 Pro joined the frontier (quesma.com) |
| 2 points by stared 17 days ago | past |
|
| | Gemini 3.7 Flash, Grok 4.6, GLM-5.3 and DeepSeek V4 Pro joined the frontier (quesma.com) |
| 5 points by stared 20 days ago | past |
|
| | Tokenflation: When "Hi" triggers 33 tool calls (quesma.com) |
| 4 points by Bluestein 30 days ago | past |
|
| | Claude Code pricing: same tokens, same model, up to 40x the price (quesma.com) |
| 31 points by jakozaur 30 days ago | past | 10 comments |
|
| | I wired 4 models together in Claude Code. It backfired 4 ways on Terminal-Bench (quesma.com) |
| 7 points by bkotrys 30 days ago | past | 1 comment |
|
| | Quantization hurts knowledge nonlinearly – Qwen3.6 27B case study (quesma.com) |
| 2 points by stared 37 days ago | past |
|
| | Kimi K3 Is Open, Opus 5 Is Good, DeepSeek V4 Flash Is Cheap: LLMs on Baba Is You (quesma.com) |
| 1 point by stared 38 days ago | past | 1 comment |
|
| | Baba Is Solved by Fable 5 and GPT-5.6 Sol, but at what cost? (quesma.com) |
| 2 points by stared 40 days ago | past | 1 comment |
|
| | Benchmarking Kimi K3, Opus 5, Grok 4.5, and Gemini 3.6 Flash on Baba Is You (quesma.com) |
| 4 points by stared 43 days ago | past |
|
| | Do Qwen 3.6 27B quantizations break the pelican? (quesma.com) |
| 9 points by stared 45 days ago | past |
|
| | Tokenflation: When "Hi" triggers 33 tool calls (quesma.com) |
| 6 points by jakozaur 46 days ago | past | 1 comment |
|
| | Baba Is Solved by Fable 5 and GPT-5.6 Sol, but at what cost? (quesma.com) |
| 4 points by agluszak 50 days ago | past |
|
| | I burned all my tokens researching how to save tokens (quesma.com) |
| 181 points by bkotrys 53 days ago | past | 211 comments |
|
| | Baba Is Solved by Fable 5 and GPT-5.6 Sol, but at what cost? (quesma.com) |
| 3 points by stared 54 days ago | past | 4 comments |
|
| | Baba Is Solved by Fable 5 and GPT-5.6 Sol, but at what cost? (quesma.com) |
| 7 points by stared 55 days ago | past | 3 comments |
|
| | The true cost of saying "Hi" to an AI agent (quesma.com) |
| 9 points by stared 63 days ago | past | 4 comments |
|
| | Qwen 3.6 27B is the sweet spot for local development (quesma.com) |
| 1192 points by stared 72 days ago | past | 759 comments |
|
| | Compare harnesses not models: Blitzy vs. GPT-5.4 on SWE-Bench Pro (quesma.com) |
| 1 point by stared 4 months ago | past |
|
| | Compare harnesses not models: Blitzy vs. GPT-5.4 on SWE-Bench Pro (quesma.com) |
| 3 points by stared 5 months ago | past |
|
| | Show HN: Reviving a 20-year-old puzzle game Chromatron with Ghidra and AI (quesma.com) |
| 28 points by stared 6 months ago | past | 9 comments |
|
| | We hid backdoors in ~40MB binaries and asked AI + Ghidra to find them (quesma.com) |
| 245 points by jakozaur 6 months ago | past | 98 comments |
|
| | We hid backdoors in binaries – Opus 4.6 found 49% of them (quesma.com) |
| 2 points by stared 6 months ago | past | 1 comment |
|
| | BinaryAudit: Can AI find backdoors in raw machine code? (quesma.com) |
| 2 points by stared 6 months ago | past | 1 comment |
|
|
| More |
