# Jais 3 goes open, Step 5 and Ling 3.1 land, Drama 3 speaks > Four open-weights model cards I read today: Jais 3 flagship, StepFun's Step 5 Preview, Ling-3.1-flash and Fish Audio's Drama 3 voice model. Published 15 Oct 2026. Coverage: 14 Oct 2026 to 15 Oct 2026. By Ali Asif. ## Items 1. **Jais-3-375B-A23B** (Inception, Open weights): The flagship of the Jais-3 family is a 375B mixture-of-experts reasoning model with 23B active parameters and a 512K context window, released under Apache 2.0 for Arabic and English. [Primary source](https://huggingface.co/inception42/Jais-3-375B-A23B) 2. **Step 5 Preview** (StepFun, Agents and coding): A 600B mixture-of-experts model with 27B active parameters and up to 1M context, aimed at agentic software engineering and finance work, with FP8 and BF16 checkpoints. [Primary source](https://huggingface.co/stepfun-ai/Step-5-Preview) 3. **Ling-3.1-flash** (inclusionAI, Open weights): A 560B hybrid-linear MoE with 25B active parameters and a 1M-token window, MIT licensed, with day-0 SGLang support and multi-token-prediction decoding. [Primary source](https://huggingface.co/inclusionAI/Ling-3.1-flash) 4. **Fish Audio Drama 3** (Fish Audio, Open weights): A 35B text-to-speech model with 3B active parameters that carries up to 16 speakers in one text, with voice cloning and tags for sounds like sighs and laughs. [Primary source](https://huggingface.co/fishaudio/drama-3) No single frontier launch today, but the open-weights shelf got heavier. I read four model cards, and three of them are sparse mixture-of-experts models north of 350 billion parameters. ## Jais 3: an Arabic and English flagship under Apache 2.0 Inception published the Jais-3 family, and the card for Jais-3-375B-A23B is the one to read. It stores 375B parameters, runs 23B per token, reasons before it answers and handles a 512K context window. The license is Apache 2.0, and the language tags are Arabic and English. The card compares it against Qwen3.5-397B-A17B, MiniMax M3, Qwen3.8-Flash-Next and NVIDIA Nemotron 3 Ultra, with Arabic open-ended generation sitting beside the usual reasoning suites. The authors say they audited some benchmarks for sample quality and report scores on the cleaned sets, and that a few benchmarks are still to come. I have not rerun any of it, so I treat the table as the lab's own claim. The reference deployment is 16 H100s across two nodes, or one 8-GPU H200 node, with vLLM and SGLang recipes. The family also includes smaller checkpoints, from 0.9B up to a 36B variant, so there is a way in without a cluster. ## Step 5 Preview and Ling-3.1-flash: two big agent models with 1M context StepFun calls Step 5 Preview its flagship for agentic work. It is a 600B mixture-of-experts model with 27B active, takes text, images and video, and reaches 1M tokens of context. The card stresses repository-level software engineering, long tool-using runs and finance. FP8 fits one H200 node, and a vLLM image is published for it. The license is custom, so read it before you build on it. inclusionAI's Ling-3.1-flash is 560B with 25B active, a hybrid-linear design with 63 linear-attention layers and 9 gated MLA layers, and a 1M-token window. It is MIT licensed. The card lists results on Terminal-Bench 4.0, SWE-bench Pro and several office and finance suites, and it mentions a coding harness the team may open-source later. The SGLang image ships on day zero with multi-token-prediction decoding. Both cards read as agent-first. The pitch is long runs with many tool calls, not a single clever answer. ## Fish Audio Drama 3: one script, up to 16 voices Drama 3 is a text-to-speech model with 35B parameters and 3B active, built for multi-character scenes and long-form audio. One text can carry up to 16 speakers through tags, long input is chunked and streamed, and the card covers voice cloning and editing a take by sending back the previous audio tokens. Inline tags such as sighs and laughs land at the position you write them. It runs on SGLang-Omni, and the license is Fish Audio's own community license, not an open-source one. For anyone building audio dramas or agent voices, this is worth a listen. ## What I'd watch next Independent Arabic evals for Jais 3 against the Qwen and MiniMax models it cites. Whether Step 5 moves from Preview to a final release. And whether the 1M-context claims on Step 5 and Ling hold up in real agent runs, not only on the cards. ## My take Four open releases in one day, and none of them is small. What I notice is that every card now leads with a serving recipe, because the weights are only useful if you can run them. Jais 3 under Apache 2.0 is the one I would test first, since Arabic reasoning is the gap I care about. ## Sources - https://huggingface.co/inception42/Jais-3-375B-A23B - https://huggingface.co/stepfun-ai/Step-5-Preview - https://huggingface.co/inclusionAI/Ling-3.1-flash - https://huggingface.co/fishaudio/drama-3 Canonical: https://heyaliasif.com/ai/2026-10-15-jais-3-step-5-ling-drama-3