Skip to content
Search models or paste a model link. Press Enter to search.

Qwen3 8B

Qwen/Qwen3-8B
Left and right arrow keys switch the vendor.
Qwen/Qwen3-8B architecture graph
Qwen/Qwen3-8B architecture graph
  • huggingface
  • AGX Thor
Labels
  • Text
  • LLM
  • Generation
Build
Embedl-optimized
Precision
W4A16
Runtime
TensorRT Edge-LLM 0.10.1 with Embedl optimizations
Context length
4,096 tokens
Run Qwen3 8B on AGX Thor

Alibaba's 8B Qwen3 model for chat, math and code, with optional step-by-step thinking.

Methodology

  • Throughput: one user, real conversations from the public OpenAssistant set, 2,048 tokens in, 128 out, greedy, no prompt reuse. Average of 20 requests after 3 warm-up requests. End-to-end is the whole request. Prompt processing is the 2,048 prompt tokens over the time to first token.
  • Power and memory: sampled on the device during those runs. Idle is the device at rest with the model loaded; energy per token is the run's energy over its generated tokens; peak memory is the device's.
  • Accuracy: full test sets, greedy, the same model files and settings as served. GSM8K 1319, IFEval 541, MMLU-Pro 1008 questions.
  • Size: the runtime container image, compressed as downloaded and unpacked on disk, and the model's files; total size is the model's files plus the container download.
  • Baseline TensorRT Edge-LLM: the original FP16 model on upstream TensorRT Edge-LLM, same settings, same device.

Files

Image Size

Model

docker.io/embedl/model-artifacts:qwen3-8b.huggingface_int4_embedl-dflash_tensorrt-edgellm.9d3a10b092e65234

7.67 GiB

Runtime

docker.io/embedl/models:runtimes.tensorrt-edgellm.arm64.2026101815014ee0

0.91 GiB

Variants

Every build of Qwen3 8B on AGX Thor.

Variant Precision Speed Peak memory Energy / token Runtime

huggingface_fp16_tensorrt-edgellm

FP16 15.1 tok/s
Throughput: End-to-end
17.41 GiB 4.11 J
TensorRT Edge-LLM 0.10.1

huggingface_int4_embedl-dflash_tensorrt-edgellm

W4A16 56.4 tok/s
Throughput: End-to-end
10.27 GiB 1.49 J
TensorRT Edge-LLM 0.10.1 with Embedl optimizations
runtimes.tensorrt-edgellm.arm64.2026101815014ee0