Qwopus3.8-27B-Flash-V2

# Fix list: Qwopus3.8-27B-Flash-V2

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/qwopus3-8-27b-flash-v2, the October 2026 research run, assessed 21 October 2026. Grade C, 61 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 Qwopus3.8-27B-Flash-V2: 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, 50 out of 100, up to 15.4 more on the total

Why it scored 50: File safety 25, safetensors and GGUF only, nothing from the repository runs. Provenance 5 of 15, the files are on a personal account with no site or organisation behind it, though every file has a checksum. Safety evaluation 0, the card says none was done. Agent guidance 5, general cautions about loops and tool reliability. Licence risk 15, plain Apache 2.0. Security programme 0, no contact or policy.

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. Documentation and upkeep, 56 out of 100, up to 6.8 more on the total

Why it scored 56: Model card 11 of 20, missing architecture, context and a cut-off. Training data 5, a description with no cut-off. Last change 20, the files went up 30 days ago. Responsiveness 10, Jackrong answered 2 of 7 discussions across the model's two repositories. Version policy 5, the first Flash release stays up without a dated note. File health 5, no LICENSE file, an experiment note in the template and open reports of looping and a grammar failure.

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.

## 3. Agent interface, 60 out of 100, up to 6.2 more on the total

Why it scored 60: Chat template 20, a Jinja file in the repository. Tool-calling format 10 of 25, only the template shows the Hermes-style JSON calls, which differ from the base's format. Trained tool calling 20, several calls a turn. Structured output 0, not documented. Reasoning control 0, the switches are documented but not where the reasoning comes back. Context 10, 262,144 tokens from the base.

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.

## 4. Reach, 85 out of 100, up to 5.8 more on the total

Why it scored 85: Memory 25 of 30, Jackrong's IQ4_XS GGUF is 15.4 GB, about 18.5 GB with a fifth added, which fits 24 GB. Runtimes 20, it keeps Qwen3.8-27B's architecture, which released vLLM, SGLang, Transformers and llama.cpp run, while Ollama's entry is the base's, not this model's. Builds 15, GGUFs at many precisions from 4-bit to 8-bit and BF16. Terms 15, Apache 2.0 through the chain. Download 10, nothing is gated. Hosted 0, no provider serves 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.

## 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.

- Jackrong is a personal account. BATCH.md's derivatives ruling speaks of a model another organisation publishes from its own organisation, so whether an individual's fine-tune counts is the founder's call
- The card says the training data includes traces derived from Claude and GPT. That's the model providers' terms, not the weights' licence, and isn't scored; the founder may want it said on the listing
- Responsiveness counts the GGUF repository with the main one, since Jackrong's GGUF is where nearly all the downloads and questions are. Counting the main repository alone gives 0
- Discussion 1 on the main repository, closed, is titled as a serious matter; its text wasn't read

## Weaknesses

- Published from a personal account, with no organisation, security contact or safety evaluation
- Training data includes traces derived from Claude and GPT, by the card's own account
- Tool calls switch to a Hermes-style JSON format that the card doesn't document
- Reasoning output location, context and cut-off aren't documented
- Reports of looping and a grammar failure wait without a reply

## 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.

- Quickest start is `llama-server -hf Jackrong/Qwopus3.8-27B-Flash-V2-GGUF:Q4_K_M --jinja`, about 17 GB, and add mmproj-F32.gguf for images
- The card's tests used Q5_K_M with the bundled MTP head and `--spec-type draft-mtp` in llama.cpp
- The template asks for tool calls as JSON inside tool_call tags, so use a Hermes-style parser (vLLM's hermes), not Qwen3.8's qwen3_coder
- For coding the card recommends enable_thinking true, reasoning_effort xhigh and temperature 0.3, which you must pass yourself, since generation_config.json says 1.0
- reasoning_effort takes xhigh (default), medium or low, and any other value makes the template raise an error
- Earlier reasoning stays in the context unless preserve_thinking is false
- Cap max_tokens, since a report says it loops more than V1

## What the review panel asked for

- a LICENSE file (2 reviews)
- add a LICENSE file
- date the version note
- document the tool format
- 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.