Latest

Every published item, in order · 9305 items
Nov 8
Tomorrow 08:00, the daily lands in your inboxTelegram channel or email. Same screening as this page, no ads.
ModelsBuildersSelected00:43
HyperFlow H3 Flash: Slimmed MiniMax-H3 Runs on Consumer GPUs

Shrinking a video generation model to 9.1GB moves real-time 768P generation from data-centre hardware into the hands of hobbyists with a single consumer GPU. Developers and indie creators can now experiment with MiniMax-H3–class video models locally instead of renting cloud capacity. GitHub access means the community can benchmark it directly against heavier open-source video generators.

Event
Gorden Sun@Gorden_SunSource
Nov 7
ModelsUS vs ChinaMulti-sourceSelected07:54
Tencent Hunyuan Hy4 tops Agent Arena as best Chinese model

Cost-per-task of US$0.57 with top-5 agentic performance makes Hy4 a strong price-performance option for agent workloads. Developers evaluating Chinese alternatives to US frontier models can compare it directly on Agent Arena's session-based leaderboard. The model is relevant to ASEAN teams building on Tencent's ecosystem, where Hunyuan integration is already common.

Event
lmarena.ai@lmarena_aiSource
ModelsBuildersMulti-sourceSelected06:51
Inkling 2 Small Ranks #5 in Code Arena: WebDev

A jump from #92 to #5 in one category suggests the small-model tier is closing the gap with frontier systems on real web-development tasks. Developers who fine-tune small models can already try it on Tinker for inference and training before the full weights drop. Once released, the open weights would make it the top US open model on this leaderboard, a direct rival to Chinese open-weight releases.

Event
lmarena.ai@lmarena_aiSource
ModelsUS vs ChinaMulti-sourceSelected04:38
DeepSeek V4.1 Flash gains 1.5x speed on GB300 NVL72

Serving-stack software, not model changes, drove most of the speedup — a signal that inference infrastructure tuning is now as decisive as raw hardware for agentic workloads. Teams benchmarking DeepSeek deployments on GB300 or comparing against NVIDIA's Dynamo + vLLM reference setup can use AA-AgentPerf numbers as a realistic baseline for agentic coding performance.

Event
Official accountArtificial Analysis@ArtificialAnlysSource
ModelsUS vs ChinaMulti-sourceSelected03:58
OpenAI's GPT-6 Astra scores 66.7% on wet-lab benchmark

The gain came with fewer reasoning tokens, so the benchmark improvement reflects better raw-data reading rather than longer thinking. Rubrics that reward partial credit and error-spotting make this a more realistic measure of lab readiness than pass/fail tests. Teams evaluating agents for scientific work can use it to compare GPT-6 Astra against GPT-5.6 Sol on messy real-world inputs.

Event
rohanpaul_ai@rohanpaul_aiSource

Only the latest 2000 items are listed; for older ones see the daily archive