# Fix list: Swift 1.5 Qwen3.8-Flash-Next From Anchor Terminal's listing at https://www.anchorterminal.com/tools/swift-1-5-qwen3-8-flash-next, the October 2026 research run, assessed 21 October 2026. Grade D, 52.3 out of 100. This is everything the published grade says the listing lacks, the biggest possible gain to the total first. It comes from the reason given for each score, the checklist each category was scored against (https://www.anchorterminal.com/benchmark/#checklist), the provenance checks, the deductions, what we couldn't check and what the review panel asked for. A fix counts at the next check, once it's public. For a coding agent working on Swift 1.5 Qwen3.8-Flash-Next: work through the items below in the product, its docs and its public pages. Each category gives the reason for its score, with the points each checklist item earned, and the checklist itself, so the gap is the items that earned less than their points. Change the product, not the wording, and keep a note of what you changed and where it's published. ## 1. Risk, 45 out of 100, up to 16.9 more on the total Why it scored 45: File safety 25, safetensors only and no code in the repository, though the card's vLLM command passes trust-remote-code. Provenance 15, the files are in the account UkisAI's site links to, with a checksum on every file. Safety evaluation 0, none published. Agent guidance 0, nothing on prompt injection, refusals or safeguards. Licence risk 5, read across the chain, Qwen's field-of-use limit and the revenue and user thresholds of both licences. Security programme 0, no way to report a flaw was found. The checklist (https://www.anchorterminal.com/benchmark/#checklist-risk): What could go wrong when you run it, and what the publisher has done about it. A higher score means less risk. - 0 to 25, file safety. 25 for weights in safetensors or GGUF only, 15 when pickle files (.bin, .pt, .pth) sit beside them, 0 for pickle only. 10 off when the documented way to load it or to build its prompts runs code from the repository (`trust_remote_code`, or the publisher's scripts in place of a template). - 0 to 15, provenance. 10 when the files come from the publisher's own organisation, the one its site or docs link to, and 5 for a checksum on every file. - 0 to 20, safety evaluation. 20 when results are published and someone other than the publisher ran them or can rerun them, 10 for the publisher's own results, 0 for none. - 0 to 15, guidance for agent use. 15 for documented prompt-injection guidance or an instruction hierarchy the model was trained to follow, 10 for a statement of what its trained refusals cover and the safeguards to add around it, 5 for general cautions, 0 for none. - 0 to 15, licence risk. 15 to start, 5 off for each of these, a use policy or field-of-use limits beyond the law, terms the licensor can change or revoke, a revenue or user threshold, and a licence that changed without a dated note or whose link is broken. Not below 0. - 0 to 10, a security programme. 5 for a way to report a flaw (security.txt, a disclosure policy or a security contact), 5 for advisories handled in public. When our reference-machine runs start, measured robustness (prompt injection with AgentDojo, over-refusal with XSTest) joins Risk. ## 2. Reach, 60 out of 100, up to 15.4 more on the total Why it scored 60: Memory 5 of 30, the smallest 4-bit build is UkisAI's IQ4_XS GGUF at 98.6 GB with the vision projector, 118.3 GB with a fifth, above the reference machine's GPU share and within 256 GB. Runtimes 20, the architecture is Qwen3.8-Flash-Next's, which released llama.cpp, vLLM, SGLang and Transformers run, while Ollama's entry is the base's. Builds 15, GGUF, INT4 and NVFP4 builds from UkisAI. Terms 10, two custom licences that allow commercial use within limits. Download 10, not gated. Hosted 0, no provider serves it, and UkisAI's research API on the card couldn't be shown to serve it, since robots.txt disallows calling it. The checklist (https://www.anchorterminal.com/benchmark/#checklist-reach): Whether an agent can run it, and where. - 0 to 30, memory. The smallest build from the publisher or a runtime's own library that keeps 4-bit precision or better, at its file size plus a fifth for the runtime and a working context. 30 for 16 GB or less, 25 for 24 GB, 20 for 32 GB, 15 for 64 GB, 10 for what fits our reference machine (the GPU share of 128 GB of unified memory), 5 for 256 GB, 0 above. - 0 to 25, runtimes. 5 for each of llama.cpp, Ollama, vLLM, SGLang, LM Studio and Hugging Face Transformers that runs the architecture in a released version, from the runtime's own code, library or documentation. A publisher's fork or an unreleased branch doesn't count. 25 at most. - 0 to 15, builds from the publisher. 15 for quantised builds at two precisions or more (FP8, 4-bit, GGUF), 10 for one, 0 when only other people's quantisations exist. - 0 to 15, terms of use. 15 for an OSI-approved licence with nothing added, 10 for a custom licence that lets anyone use it commercially for free below a threshold or under a use policy, 5 for research or non-commercial use only, 0 for no licence or unclear terms. - 0 to 10, download. 10 when the files download with no account and no terms to accept, 5 for a click-through that needs an account, 0 when a person approves each request. - 5, hosted too. The publisher's own API serves it, or at least two inference providers do, so an agent without the hardware can still call it. When our reference-machine runs start, speed on that machine (tokens per second and time to first token at the build that fits) joins Reach. ## 3. Documentation and upkeep, 40 out of 100, up to 9.2 more on the total Why it scored 40: Model card 5 of 20, intended use and context only. Training data 5, a general description with no cut-off. Last change 20, released 27 days ago. Responsiveness 0, UkisAI answered none of 3 discussions. Version policy 5, a first release, renamed without a note. File health 5, the re-serialised tokenizer splits text differently from the repository's own tokenizer config and the base. The checklist (https://www.anchorterminal.com/benchmark/#checklist-upkeep): - 0 to 20, the model card. Intended use, limitations, architecture, parameters, context length, languages and knowledge cut-off. 20 for all seven, 3 off for each that's missing. - 0 to 20, training data. 20 for the data published or listed with sources and sizes, 15 for a data card that names the sources or the mix, the cut-off and the filtering, 10 for a general description with a cut-off, 5 without one, 0 for none. - 0 to 20, the last change to the files, the template or the reference code. 20 within 30 days, 15 within 90, 10 within 180, 5 within a year, 0 older. - 0 to 20, responsiveness. The publisher's replies to the 40 most recent questions and reports in the repository's discussions or issues. 20 for most, 10 for some, 5 for a few, 0 for none. - 0 to 10, version policy. 10 for earlier versions kept in their own repositories with a dated note when one is replaced, 5 for earlier versions kept without notes or a first release, 0 when files were replaced in place without a note. - 0 to 10, file health. 10 when config, tokenizer, template and card agree and no report of a broken file waits without a reply, 5 when one does, 0 when the files contradict the card. ## 4. Agent interface, 60 out of 100, up to 6.2 more on the total Why it scored 60: Chat template 20, the base's Jinja file. Tool-calling format 10 of 25, the format is only in the template and the card's two commands name different parsers. Trained tool calling 20, several calls a turn and agentic post-training. Structured output 0, not documented. Reasoning control 0, the card shows how to set reasoning_effort but not how to turn thinking off or where the reasoning comes back. Context 10, 262,144 tokens native. The checklist (https://www.anchorterminal.com/benchmark/#checklist-interface): What a model and a developer need to talk to it. Speech, image and other models with no chat interface read the tool-calling and structured-output lines for what the model takes and returns (voices, timestamps, streaming, typed outputs), and the note says how. - 0 to 20, the chat template. 20 for a template in the repository (Jinja, in the tokenizer config or its own file), 10 when the prompt format is only given as code you run. - 0 to 25, the tool-calling format. 25 when the publisher documents how tools are declared and how calls come back, with the parser to use in a named runtime, 10 when only the template or a runtime's own recipe says it. - 0 to 20, tool calling the model was trained for. 20 with several calls in one turn, 10 for one call at a time. - 0 to 15, structured output. 15 when JSON or JSON Schema output is documented for the model with how to ask for it, 5 when the card claims it without saying how. - 10, reasoning control. Thinking that can be turned off or budgeted per request, with where the reasoning text comes back. - 0 to 10, context. 10 for 128K tokens or more trained, 5 for 32K to 127K. Context reached only by scaling the position encoding counts one band down. ## What we couldn't check What we couldn't read counted as absent. Publishing it on a page a plain HTTP fetch can read (not only in a browser) lets the next check count it. - Whether UkisAI trained with the re-serialised tokenizer (so its pattern is what the weights expect) or the change is accidental isn't stated. The NOTICE only says the pattern differs from the base's - Runtimes rest on the qwen3-8-flash-next listing's check of released code for qwen4_exp. The card itself still asks for builds with Flash-Next support and an SGLang Docker image - The card's metadata says gated true while the Hub serves the files ungated - UkisAI's research API at ukisai.com/api/flash-next/v1 wasn't called, since ukisai.com's robots.txt disallows /api/, so hosted is 0 on the checklist's reading that the publisher's API must be shown to serve it. If it does, hosted is 5 ## Weaknesses - Two licences to satisfy, UkisAI's US$1M revenue limit and Qwen's separate licence for model-as-a-service businesses - The smallest 4-bit build needs about 118 GB, beyond a 128 GB unified-memory machine's GPU share - tokenizer.json's pre-tokenizer differs from the base's and from its own tokenizer config - No reply from UkisAI in any discussion, and no safety results, agent guidance or security contact - The card's two serving commands name different tool parsers, and reasoning control and JSON output aren't documented ## What costs an agent a turn today The notes we give agents before they call it. Each one is a workaround an agent shouldn't need. - On a multi-GPU node use `vllm serve ukisai/Swift1.5-Qwen3.8-Flash-Next --tensor-parallel-size 8 --max-model-len 262144 --reasoning-parser qwen3 --enable-auto-tool-choice --tool-call-parser qwen3_coder`, the parser the base's recipes and the card's SGLang command use - The card's vLLM command passes --trust-remote-code and the old repository name, neither is needed - On our reference machine only UkisAI's GSQ-RCO IQ3_S GGUF (84.7 GB with the projector) fits, below 4-bit - Set reasoning effort per request with `chat_template_kwargs` `reasoning_effort` xhigh (default), medium or low. UkisAI's own runs lose about 2.5 points at medium and low - If text with combining marks matters, compare tokenisation against the base's tokenizer.json before trusting outputs - Check both licences before commercial use, UkisAI's stops at US$1M gross annual revenue and Qwen's needs a separate licence for model-as-a-service businesses ## What the review panel asked for - note the rename - state which tokenizer pattern was trained - published licence prices - state which tokeniser pattern is intended - publish safety results - a security contact ## When it's done Send what changed and where it's published as a dispute (https://www.anchorterminal.com/builders/#disputes, or `POST https://www.anchorterminal.com/api/v1/contact` with `"kind": "dispute"`). Disputes are answered in public, and the listing is checked again by the same checklist. Paying for an audit or a listing claim changes nothing here.