batch-infer-sglang
Batch chat completions with SGLang
Runs an OpenAI batch file of /v1/chat/completions (or /v1/completions) requests through an SGLang server started inside the job and writes one result line per request, in the same shape as batch-infer-vllm. SGLang's RadixAttention reuses the KV cache of prefixes the lines share, so a batch with one long system prompt or few-shot block does that prefill once. Default model Qwen3-4B-Instruct-2507 (Apache-2.0), downloaded from Hugging Face at start.
- Status
- Available
- GPU
- NVIDIA A10G · 24 GB
a10g-24gb-x1 - Default runtime
- 1 h maximum
- Publisher
- Unify Compute
What it reads and writes
Input /mnt/in | Directory containing requests.jsonl: one OpenAI batch line per request with url /v1/chat/completions or /v1/completions; the model field is replaced with MODEL |
|---|---|
Output /mnt/out | results.jsonl: one OpenAI batch result line per request, in input order, matched by custom_id; a line SGLang could not answer carries error instead of response |
What you can change
A submit may set: GPU (SKU) (sku), Cloud placement (providers), Environment variables (spec.env), Maximum runtime (spec.max_runtime_seconds). The image and the declared inputs and outputs are fixed: a different image is a different template.
MODEL | Qwen/Qwen3-4B-Instruct-2507 |
|---|---|
MODEL_REVISION | cdbee75f17c01a7cc42f958dc650907174af0554 |
MAX_MODEL_LEN | 8192 |
CONCURRENCY | 64 |
Run it
With a key in $UNIFY_KEY (see the quick start), give each input an https:// URL, public or presigned, that ends in the file name above:
curl -s https://api.compute.unifyai.us/compute/v1/jobs \
-H "Authorization: Bearer $UNIFY_KEY" \
-H "Content-Type: application/json" \
-d '{
"template": "batch-infer-sglang",
"spec": {
"inputs": [
{
"source": "https://YOUR-BUCKET.s3.amazonaws.com/requests.jsonl",
"mount": "/mnt/in"
}
]
}
}'The job is billed for the GPU time it uses at the SKU's rate, the same as a job you write yourself.
GPU job execution is turned on per account while it rolls out. Create an account, then ask us to enable it as the quick start describes.
Where the image comes from
| Image | lmsysorg/sglang:v0.5.19-cu129-runtime |
|---|---|
| Digest | sha256:fe08f6f52bea380415792849a703d92279c86320f8c39283bcd04043f8b63c41 |
| Licence | Apache-2.0 |
| Source | https://hub.docker.com/r/lmsysorg/sglang (tag v0.5.19-cu129-runtime, pushed 2026-09-04, the newest CUDA 12.9 build; multi-arch index read from registry-1.docker.io on 2026-11-06; LICENSE at github.com/sgl-project/sglang/blob/v0.5.19/LICENSE) |
The digest is what runs. A tag can be re-pushed; a digest cannot, so what was reviewed is what your job pulls.