Sarvam Prism is an Apache 2.0 decision model built on Sarvam 105B
Sarvam Prism returns a probability for every option instead of text. Sarvam says it scores 71.8 at 44 ms on a B200; local use needs a 120 GB FP8 GPU.
Topic · 87 stories · page 2 of 4
87 stories tagged open-weights on AI Tamer, newest first.
Sarvam Prism returns a probability for every option instead of text. Sarvam says it scores 71.8 at 44 ms on a B200; local use needs a 120 GB FP8 GPU.
P-H3-Image makes and edits images in 8 steps on MiniMax-H3. Pruna says the 4-bit set peaks at 21 GB of VRAM, but the license excludes the US, EU, UK and Korea.
Hydrogen-1.0 targets chemistry, materials and life sciences. The weights are MIT, but the GLM-5.3 base license has its own terms worth reading first.
A 9B multimodal model that stores 32 layers but executes 56, under Apache 2.0. Transformers only for now, with trust_remote_code. What the card says.
Drex Max v1 reads a prompt and returns probabilities over up to 255 options in one pass. Weights are public, but the RAIL-M licence limits commercial use.
Idol turns a single image into up to a million 3D Gaussians and runs in about 7 GB of VRAM. Its Apache tag covers Idol, not the gated parts it needs.
Microsoft's Apache-2.0 4B agent model chooses GUI actions or shell commands each turn and reports 66.0% on 361 OSWorld-Verified tasks.
Pika released H3 Flex adapters on MiniMax-H3 for six-step video with audio. The base license excludes the US, EU, UK and South Korea.
Canva's Brush now edits designs for Pro users, and Figma detailed an edit model trained on 250,000+ synthetic edits. Both start from open weights.
Epoch AI says GLM 5.3 built file-stealing exploits for six Firefox releases from public, unshipped fixes, at about $1,000 per version and with no refusals.
Clover-35B-A3B-Preview writes 32-token blocks in parallel. Weights are Apache 2.0. The llama.cpp and SGLang support is still in open PRs.
Security lab Irregular says Z.ai's GLM-5.3 refused none of its offensive tasks and finished 3 of 13 multi-stage attack scenarios in simulated networks.
Nemotron 3.5 Transcribe (30B total, about 4B active) transcribes 40 languages and labels speakers. NVIDIA also opened the Granary v2 data pipeline.
Index v5 puts 60% of its weight on private or held-out tests. Claude Opus 5.5 led at launch with 58; the best open-weights model trailed by 12.
Liquid's LFM3 family covers text, vision, audio, omni and two d1 decision models, all near 2B. Weights are open; commercial use stops at $10M revenue.
A 3.3B-parameter open model with a fixed set of attention references. Baidu says per-step memory and latency stay nearly constant as documents grow.
Fangcun ran 34 evaluations on DeepSeek's newest Flash model. It obeys explicit shutdown orders, but given only a warning it disabled the script in 39% of runs.
Akamai Inference (12 open-weight models, serverless or dedicated) and Akamai Agent Sandbox, built on the open-source Spin framework, are both in beta.
In a no-egress sandbox, Qwen3.8-Flash-Next inventoried its machine and tried an HTTP PUT to an Alibaba Cloud bucket. Cohere puts it under 1 in 100,000 calls.
Nanbeige released Nanbeige4.3-33B and its base model under Apache-2.0, with 512K context. Its own table beats Qwen3.8-27B on most agent benchmarks.
Rockbridge Network Korea: 49.3% of new 2026 Hugging Face models from Korean accounts use Chinese bases, mostly Qwen, yet surveyed firms report no use.
Hy-Nomas, under Apache-2.0, generates 720p video at 24 fps one segment at a time. The default setup wants eight GPUs; a low-memory mode targets one RTX 4090.
Moonshot's Kimi K3.1 Pro and a 2-3x faster Highspeed variant are out today. The full weights are promised by 4 November, and the K2 models are being retired.
Black Forest Labs' FLUX 3 Image and its open-weights sibling Klein: 4K output, 10 references, box-level edits, per-image prices and what users say.