Hackernews posts about Quesma

RSS feed for Quesma

01
newBenchmarking Qwen3.8 27B quantizations: 4-bit holds up, 1-bit collapsesquesma.com
▲ 286stared 27d 💬 139
02
newRTK reports token savings, but our cost benchmarks disagreequesma.com
▲ 170michalwarda 24d 💬 84
03
newHackathons in the vibe coding era: our token economics experimentquesma.com
▲ 3jakozaur 14d 💬 2
04
newHackathons in the vibe coding era: our token economics experimentquesma.com
▲ 3jakozaur 20d discuss
05
newAgent session transcripts are precious, keep themquesma.com
▲ 2hugodutka 2h discuss
06
newAgent session transcripts are precious, keep themquesma.com
▲ 2stared 4d discuss
07
newGPT-6 Astra solves puzzlesquesma.com
▲ 2stared 21d discuss
08
newKnowledge vs. wisdom: asking AI "What mushroom is that?"quesma.com
▲ 2stared 26d discuss
09
newIlha Da Queimada Grandeen.wikipedia.org
▲ 6thunderbong 15d discuss
10
newAlibaba, Datadog, and Quesma Join Forces on Go Compile-Time Instrumentationopentelemetry.io
▲ 37darccio 619d 💬 4
11
newShow HN: Ggplot2 chart playground – in the browser with WebAssemblygithub.com
▲ 4stared 384d discuss
12
newQwen 3.6 27B is the sweet spot for local developmentquesma.com
▲ 1192stared 97d 💬 759
13
newWe hid backdoors in ~40MB binaries and asked AI + Ghidra to find themquesma.com
▲ 245jakozaur 225d 💬 98
14
newTau² benchmark: How a prompt rewrite boosted GPT-5-mini by 22%quesma.com
▲ 197blndrt 383d 💬 65
15
newI burned all my tokens researching how to save tokensquesma.com
▲ 181bkotrys 78d 💬 211
16
newThe security paradox of local LLMsquesma.com
▲ 160jakozaur 348d 💬 87
17
newCompileBench: Can AI Compile 22-year-old Code?quesma.com
▲ 148jakozaur 378d 💬 65
18
newBenchmarking OpenTelemetry: Can AI trace your failed login?quesma.com
▲ 144stared 249d 💬 79
19
newMushroom hunting with LLMs: what can go wrong?quesma.com
▲ 56stared 32d 💬 76
20
newClaude Code pricing: same tokens, same model, up to 40x the pricequesma.com
▲ 31jakozaur 55d 💬 10
21
newShow HN: Reviving a 20-year-old puzzle game Chromatron with Ghidra and AIquesma.com
▲ 28stared 212d 💬 9
22
newTau² Benchmark in Action: Early Results and Key Takeawaysquesma.com
▲ 16luciesim 399d discuss
23
newBenchmarking Qwen3.8 27B quantizations: 4-bit holds up, 1-bit collapsesquesma.com
▲ 15stared 39d 💬 2
24
newAI Tokenomics: What AI tokens cost and where they're wastedgithub.com
▲ 13bkotrys 71d discuss
25
newThe true cost of saying "Hi" to an AI agentquesma.com
▲ 9stared 88d 💬 4
26
newDo Qwen 3.6 27B quantizations break the pelican?quesma.com
▲ 9stared 70d discuss
27
newBaba Is Solved by Fable 5 and GPT-5.6 Sol, but at what cost?quesma.com
▲ 7stared 81d 💬 3
28
newI wired 4 models together in Claude Code. It backfired 4 ways on Terminal-Benchquesma.com
▲ 7bkotrys 56d 💬 1
29
newOutside of the bubble, AI is Black Mirrorquesma.com
▲ 7stared 303d discuss
30
newWe hid backdoors in binaries – Opus 4.6 found 49% of themquesma.com
▲ 6stared 237d 💬 2
31
newTokenflation: When "Hi" triggers 33 tool callsquesma.com
▲ 6jakozaur 71d 💬 1
32
newPrompts are (not) the new source codequesma.com
▲ 6stared 269d 💬 1
33
newMigrating CompileBench to Harbor: standardizing AI agent evalsquesma.com
▲ 6stared 286d discuss
34
newA postmortem on our $2.5M database gateway: lessons from pilot purgatoryquesma.com
▲ 6jakozaur 334d discuss
35
newDon't Let Apache Iceberg Sink Your Analytics: Practical Limitations in 2025quesma.com
▲ 6jakozaur 500d discuss
36
newClickHouse Cloud Pricing Change in January 2025: A Price Hike with Many Tweaksquesma.com
▲ 6jakozaur 599d discuss
37
newWe hid backdoors in binaries – Opus 4.6 found 49% of themquesma.com
▲ 5stared 236d 💬 2
38
newNano Banana Pro: raw intelligence with tool usequesma.com
▲ 5stared 315d 💬 2
39
newAI for coding is still playing Go, not StarCraftquesma.com
▲ 5stared 355d 💬 2