`Qwen/Qwen3.5-4B`

apache-2.0 1 launch

image-text-to-text · 4.66B parameters · 8.7 GiB of weights at revision 851bf6e8. Runs on a single GPU when the source verified it.

download the card performance the weights on HF

Did this card launch? The count is real — it comes from launches the kit recorded against this path, not from a widget anywhere.

1checkpoints

1published launches

1GPU minimum

2named gaps

The identity, read from the HF API at the revision the card pins.
weight path	`Qwen/Qwen3.5-4B`
revision	`851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a`
license	apache-2.0
gated	no
parameters	4.66B
weight bytes	8.7 GiB
modality	image-text-to-text
HF downloads	7.0M at capture
last modified	2026-03-02T00:52:52.000Z

The checkpoints.

Every servable variant in or beside the repo. resident_gib is what lands in VRAM after load; host_gib is what stays on the host and never touches the card.

name	format	location	resident	host tier	gated
`root`	bf16	repo root	8.7 GiB	—	no

root — 2 weight file(s)

The published launches.

One entry per engine × checkpoint with a command somebody published. Each cites the page it was read from — the card copies it, it does not invent it.

vllm `root`

vllm/vllm-openai-xpu:latest

argv — what a wrapper execs

vllm serve Qwen/Qwen3.5-4B --trust-remote-code --tensor-parallel-size 1 --enforce-eager --enable-auto-tool-choice --tool-call-parser qwen3_coder --reasoning-parser qwen3

as the source printed it

vllm serve Qwen/Qwen3.5-4B \
  --trust-remote-code \
  --tensor-parallel-size 1 \
  --enforce-eager \
  --enable-auto-tool-choice \
  --tool-call-parser qwen3_coder \
  --reasoning-parser qwen3

validated on: xeon6, arc_pro_b60, arc_pro_b70
env: VLLM_USE_RUST_FRONTEND=1
source: vllm-recipes

Every floor a box must clear.

A launch that fails on the third requirement wasted the first two. These are the requirements, as data — the fit gate reads them instead of a paragraph.

The hardware block, verbatim from the card.
GPU models	none validated
architectures	—
min GPUs	1
TP rule	widest published TP is 1
min VRAM / GPU	—
min total VRAM	—
min disk	60 GiB
min host RAM	—
interconnect	not required
CUDA min	—
driver min	—
context tokens	—

The 2 things this card could not answer.

First-class, not swallowed. Every entry is a field a source did not answer — which makes it a launch nobody has done, and therefore a unit of work rather than a defect.

envs.sglang no SGLang cookbook page matched this repo

hardware.validated the only verified targets are non-GPU: arc_pro_b60, arc_pro_b70, xeon6 — no GPU family was verified for this repo

The file.

The card is a YAML document with a published JSON Schema. View it, download it, diff it against the next capture — and point one at it.

download

launch-card.yaml card.json the schema

Same bytes the collector wrote. The JSON is the same card, for a machine that would rather not parse YAML.

with one

one launch --card Qwen/Qwen3.5-4B/launch-card.yaml

one follows the argv above, checks the floors before the paid clock, and on a failure changes exactly one field and retries — recording every attempt against this path.

Where every field came from.

The card is assembled, not authored. These are the sources, so a reader can check the claim rather than trust it.

source	url
hf-api	`https://huggingface.co/api/models/Qwen/Qwen3.5-4B`
vllm-recipes	`https://recipes.vllm.ai/Qwen/Qwen3.5-4B.json`

Captured 2026-10-31T17:28:20Z by hf-api + vllm-recipes + sglang-cookbook. Schema onefrontier.launch-card/v1.