onefrontieronefrontier.ai

Qwen/Qwen3.8-27B

apache-2.0 1 launch

image-text-to-text · 27.78B parameters · 51.7 GiB of weights at revision 1d4bf0f2. 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 pathQwen/Qwen3.8-27B
revision1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0
licenseapache-2.0
gatedno
parameters27.78B
weight bytes51.7 GiB
modalityimage-text-to-text
HF downloads7.6M at capture
last modified2026-08-14T15:00:01.000Z

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 51.7 GiB — no

root — 18 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:qwen38
argv — what a wrapper execs 
vllm serve Qwen/Qwen3.8-27B --tensor-parallel-size 1 --enable-auto-tool-choice --tool-call-parser qwen3_coder --reasoning-parser qwen3 --mm-encoder-tp-mode data
as the source printed it
vllm serve Qwen/Qwen3.8-27B \
  --tensor-parallel-size 1 \
  --enable-auto-tool-choice \
  --tool-call-parser qwen3_coder \
  --reasoning-parser qwen3 \
  --mm-encoder-tp-mode data
validated on
gb300, b200, rtx_5090, rtx_5090_2x, rtx_pro_6000, dgx_spark_gb10, ascend_950pr, 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 modelsb200, gb300, rtx_5090, rtx_5090_2x, rtx_pro_6000
architecturesblackwell
min GPUs1
TP rulewidest published TP is 1
min VRAM / GPU—
min total VRAM—
min disk112 GiB
min host RAM—
interconnectnot 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 SGLang cookbook page exists but its launch command is generated by the page's Deploy widget, not printed in the markdown; the page documents these flags: --mamba-full-memory-ratio, --speculative-num-draft-tokens, --speculative-dspark-block-size, --enable-linear-replayssm-spec, --mamba-ssm-dtype, --max-mamba-cache-size, --max-running-requests, --mem-fraction-static, --max-total-tokens, --kv-cache-dtype, --attention-backend, --device, --speculative-algorithm, --tp-size...
hardware.npu recipes name non-GPU accelerators too: arc_pro_b60, arc_pro_b70, ascend_950pr, dgx_spark_gb10 (not carded)

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/Qwen3.8-27B/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.

sourceurl
hf-api https://huggingface.co/api/models/Qwen/Qwen3.8-27B
vllm-recipes https://recipes.vllm.ai/Qwen/Qwen3.8-27B.json
sglang-cookbook https://docs.sglang.io/cookbook/autoregressive/Qwen/Qwen3.8-27B.md

Captured 2026-10-31T17:28:18Z by hf-api + vllm-recipes + sglang-cookbook. Schema onefrontier.launch-card/v1.