LLaDA 2.2-flash shows broad benchmark gains over Ling-2.6-flash
omarsar0 · x · 2026-07-26
The benchmark chart highlights LLaDA 2.2-flash’s gains across agentic and coding tasks.
- SWE-bench Verified: 49.28 vs 31.88 in the highlighted comparison.
- The same graphic also shows strong lead margins on SWE-bench Pro, SWE-bench Multilingual, τ²-Bench, BFCL-v4, MCP-Atlas, PinchBench, Claw-Eval, and related tasks.
More from Models
- Benchmark scores drop from 89% to 19% on new evals — how benchmaxxing breaks leaderboard trust — airesearch12 · 2026-09-11
- Developer Building a Unified Leaderboard of All Model Benchmark Scores — airesearch12 · 2026-09-11
- Rumor claims Kimi faked performance by serving Claude; DeepSeek new model surprises in evals — realsohamparekh · 2026-09-11
- GPT-5.6 writes well but is instantly forgettable, user complains — BasedRaddka · 2026-09-11
- Opus Refuses Protein Research Codebase Over 'Safety' Concerns, Dev Considers Rolling His Own — josephdviviano · 2026-09-11
- User Hails Unconfirmed 'DeepSeek 4.1 Flash' as an Inflection Point in LLMs — himanshustwts · 2026-09-11