Fetching from the wire…
01
02
03
04
05
06
07
08
Research2026-09-26 · source-backed
arXiv 2609.28765 trained Qwen3.5-0.8B with GRPO and an interleaved Wikipedia search tool on MuSiQue, varying only the reward shape across three seeds. Best run reached 0.352 average exact match on a seven-benchmark suite against a 0.092 untrained floor. The Search-R1-style exact-match-only reward was worst at every seed, including on exact match itself. No distillation needed.
Each link below shares sources, entities, or timing with this story.
New batching algorithms enable ~7x, up to 12x+, longer-context GRPO training with no accuracy or speed penalty versus optimized FA3 and chunked-loss setups (Unsloth Docs). Qwen3-8B GRPO reaches 110K context on one 80GB H100 via vLLM plus QLoRA. For solo builders doing reasonin...
Alibaba released Qwen3.6-27B on April 22. Dense architecture. Open weights. 77.2% on SWE-bench Verified, within 3.7 points of Claude Opus 4.6. On SkillsBench, it scores 48.2% versus its own 397B MoE predecessor's 30.0%. That's a 77% improvement with 14.8x fewer parameters. Let...
Two specialized RL agents — Memory Manager (ADD/UPDATE/DELETE operations) and Answer Agent — fine-tuned with PPO and GRPO. With only 152 training QA pairs, outperforms baselines across three benchmarks. Directly applicable to persistent agent memory systems. arXiv 2508.19828
Claims up to 2x faster generation, and adds fine-tuning of both MoE models on text or image datasets on Apple Silicon via MLX (release). Follow-up turns in long Qwen chats on Mac are reported up to 30x faster, MLX models now use full context size, and GLM-5.3 MLX fine-tunes ex...
Unsloth's UD-Q2_K_XL (78.9 GB) plus a 358,400-token slot via YaRN from the native 262,144 with fp16 KV fits under the default 96GB Metal wired limit, no sysctl hack. Cold prefill runs 1,561 t/s at 5.6K context down to 318 t/s at 111K; a normal incremental turn is 77-854 t/s ou...
Released August 27 with GGUFs for both, claiming 5x faster inference for RAM offloading, working repeated compaction, chats that recover after disconnects instead of losing the reply, and memory estimates shown before a load (GitHub). That's roughly a 24-hour turnaround from t...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.