AI generates this pattern because it's the shortest path from 'fetch data' to 'render table.'

AlphaEvolve improved the efficiency of Google Spanner by refining its Log-Structured Merge-tree compaction heuristics. This optimization reduced 'write amplification'—the ratio of data written to storage versus the original request—by 20%.

We reverted this change on April 7 after users told us they'd prefer to default to higher intelligence and opt into lower effort for simple tasks.

Agents and CDC streams are powerful together because they split the work well.

The fix is not smarter prompts. It is software built to meet agents halfway.

The extra tokens bought something measurable. +5pp on strict instruction-following. Small. Real. So: is that worth 1.3–1.45x more tokens per prompt?

The standard autoresearch loop (brainstorm from code, run experiments, check metrics) works when the optimization surface is visible in the source. The Liquid results prove that. But for problems where the codebase doesn't contain enough information to generate good hypotheses, giving the agent access to papers and competing implementations changes what it tries.

Without experience with compiler behavior, the agent couldn't have predicted which 'optimizations' the compiler would already handle.

Coding agents working from code alone generate shallow hypotheses. Adding a research phase — arxiv papers, competing forks, other backends — produced 5 kernel fusions that made llama.cpp CPU inference 15% faster.

Total cost: ~$29 ($20 in CPU VMs, $9 in API calls) over ~3 hours with 4 VMs.

The agent fused them into one: for (int i = 0; i < nc; i++) { wp[i] = sp[i] * scale + mp_f32[i]; }

The same task on full Codex took ~5× longer.

GLM-5.1 did not plateau after 50 or 100 submissions, but continued to find meaningful improvements over 600+ iterations with 6,000+ tool calls, ultimately reaching 21.5k QPS—roughly 6× the best result achieved in a single 50-turn session.

In 23 months, the same capability that needed 1.8 trillion parameters now fits in 4 billion parameters. A 450x compression.

Artificial Analysis has also positioned Gemini 3.1 Flash TTS within its 'most attractive quadrant' for its ideal blend of high-quality speech generation and low cost.

After compressing, the model again extends its solutions to achieve stronger performance.

For higher-interactivity scenarios, execution time for MoE models is bound by expert weight load time. By splitting, or sharding, the experts across multiple GPUs across NVL72 nodes, this bottleneck is reduced, improving end-to-end performance.