AutoGineer
Generated banner for Jetson LLM Benchmark
Generated logo for Jetson LLM Benchmark

Jetson LLM Benchmark

NVIDIAOfficialCreated by Jetson TeamDistributed by NVIDIA

  • Skill
  • Tool integration
  • Software development
  • Apache-2.0
  • Medium risk
  • Has scripts
Download ZIP

From the author

Benchmark Jetson LLM/VLM serving performance across vLLM, llama.cpp, and Ollama with structured JSON output.
— the description by Jetson Team

AutoGineer summary

Written by AutoGineer from the skill’s own files, not by the author

This skill provides a standardized workflow for measuring LLM latency and throughput on NVIDIA Jetson hardware. It includes wrapper scripts for vLLM, llama.cpp, and Ollama that handle environment-specific requirements like container selection, warmup passes, and hardware-aware metric collection. The output is a consistent JSON format, enabling agents to compare performance across different models, runtime configurations, and power modes.

What it does

  • Provides wrapper scripts to benchmark vLLM, llama.cpp, and Ollama runtimes on Jetson devices.
  • Ensures consistent JSON output for performance metrics including TTFT, ITL, TPOT, and throughput.
  • Automatically detects and reports Jetson hardware details like SKU, L4T version, and container environment.
  • Includes built-in warnings for thermal throttling, power mode issues, or background GPU usage.
  • Handles runtime-specific requirements such as warmup passes and concurrency sweeps.

When to use it

  • When you need to quantify the performance of an LLM deployment on a Jetson device.
  • To compare performance before and after applying memory tuning or configuration changes.
  • When you need reproducible, machine-readable benchmark data for model selection.
  • To diagnose performance regressions following system updates or model changes.

Advantages

  • Standardized JSON output allows for automated comparison and diffing of benchmark runs.
  • Reduces manual effort by handling container selection and runtime-specific benchmark logic.
  • Provides Jetson-specific context and warnings that generic benchmarking tools lack.
  • Encourages controlled testing by enforcing consistent input/output lengths and power modes.

Usage examples

“Benchmark the Llama-3 model running on my local vLLM server.”
The agent will invoke scripts/bench_vllm.sh with the appropriate model ID and concurrency settings, then return the structured JSON performance report.
“How do I measure the performance of a GGUF model on my Jetson Orin?”
The agent will provide the command for scripts/bench_llama_cpp.sh, explaining how to specify the model path and GPU layers for optimal results.

Install

  1. Download the skill as a ZIP:

    https://autogineer.com/agent-skills/nvidia/jetson-llm-benchmark/download

  2. Unzip it into the cross-client skills folder of your project:

    mkdir -p .agents/skills && unzip jetson-llm-benchmark.zip -d .agents/skills

Documentation by Jetson Team

Jetson LLM Benchmark

Reproducible Jetson benchmarks with structured JSON output so an agent can compare runs. Encodes the workflow from the Jetson AI Lab GenAI Benchmarking tutorial.

Purpose

Measure deployed LLM latency and throughput on a Jetson target using the correct runtime-specific benchmark wrapper. Use the JSON output to compare models, runtime flags, power modes, and before/after tuning changes.

Prerequisites

  • Run on the Jetson device that hosts the model runtime.

  • For vLLM, start the OpenAI-compatible vLLM server first and know the served model ID.

  • For Ollama, ensure the Ollama daemon is reachable at --endpoint and the named model is already pulled.

  • For llama.cpp/GGUF, provide a readable .gguf model path on the host.

  • Put the device in the intended power mode before measuring. MAXN is preferred for comparable performance numbers.

Available Scripts

ScriptPurposeArguments
scripts/bench_vllm.shRuns vllm bench serve against a running OpenAI-compatible vLLM server.--model, --endpoint, --concurrency, --input-len, --output-len, --num-prompts, --no-warmup, --container, --native.
scripts/bench_llama_cpp.shRuns llama-bench for a local GGUF model through the Jetson-appropriate NVIDIA-AI-IOT llama.cpp container.--model, --n-prompt, --n-gen, --n-gpu-layers, --threads, --container.
scripts/bench_ollama.shBenchmarks a local or containerized Ollama daemon through the /api/generate REST API.--model, --endpoint, --num-prompts, --input-len, --output-len, --no-warmup.

If your agent runtime supports run_script, invoke the selected wrapper directly with the user-provided model identifier or local model path, then summarize the returned JSON. Otherwise run the wrapper with bash {baseDir}/scripts/<wrapper-name> ....

Instructions

Always use the matching wrapper script for the runtime — do not call the underlying vllm bench serve, llama-bench, or curl against /api/generate by hand:

  • vLLM → scripts/bench_vllm.sh (required for the vLLM path)

  • llama.cpp / GGUF → scripts/bench_llama_cpp.sh (required for the GGUF path)

  • Ollama → scripts/bench_ollama.sh (required for the Ollama path)

These wrappers handle warmup, the NVIDIA-AI-IOT container selection, and JSON emission. Calling the underlying tool directly will not satisfy the output contract below.

For "how do I benchmark/measure" questions, first run the matching wrapper with --help to verify the exact options, then answer with the wrapper command. Do not run a full benchmark unless the user asks you to execute it or the required server/model path is already confirmed.

Expected Workflow

Pick exactly one wrapper based on the runtime the user named, and invoke that wrapper with --help before composing the answer. Do not merely mention the script name. If the runtime does not execute scripts relative to the skill directory, use {baseDir}/scripts/<wrapper-name>.

  • Existing vLLM OpenAI-compatible server at localhost:8000: {baseDir}/scripts/bench_vllm.sh --help, then show a command using --concurrency 1,8 and the served model ID.

  • llama.cpp / GGUF / llama-server: {baseDir}/scripts/bench_llama_cpp.sh --help, then show a command for the GGUF model path and report that prompt/generation speed maps to TTFT, ITL/TPOT, and throughput.

  • Ollama: {baseDir}/scripts/bench_ollama.sh --help, then show a command with --model <ollama-tag>. Do not use vLLM or llama.cpp wrappers for Ollama.

When to use

  • "Benchmark / measure / compare X on this Jetson."

  • After jetson-llm-serve to actually quantify the deployment.

  • Before/after applying flags from jetson-inference-mem-tune to confirm the change helped.

Three paths — pick by runtime

A. vLLM (preferred for parity with how things are served)

Server must already be running (use jetson-llm-serve). Run bench_vllm.sh:

scripts/bench_vllm.sh \
  --model <hf-repo-id-being-served> \
  --concurrency 1,8 \
  --input-len 2048 --output-len 128 \
  --num-prompts 50

Uses the Jetson-appropriate benchmark client path: upstream vLLM 0.20+ container vllm/vllm-openai:latest on Thor and Orin JetPack 7.2 / L4T r39+, or the NVIDIA-AI-IOT vLLM benchmark container ghcr.io/nvidia-ai-iot/vllm:latest-jetson-orin on older Orin. Pass --native only when host-native vLLM is already installed and validated. It runs against http://localhost:8000/v1. Always do a warmup pass first (~10 prompts, discarded) before the measured run — Jetson has cold caches and JIT'd kernels.

B. Ollama (for models served by a running Ollama daemon)

No benchmark container needed. Uses Ollama's /api/generate REST API directly — timing data (TTFT, ITL, throughput) comes from the response JSON, so no --verbose parsing is required.

Prerequisite: the Ollama daemon must be reachable at --endpoint (default http://localhost:11434). This works whether Ollama is installed natively or running in a container that exposes that port. If the daemon is not running, the script will tell you whether Ollama is installed but stopped (ollama serve to fix) or not installed at all (install instructions printed). Run bench_ollama.sh (do not roll your own curl against /api/generate):

scripts/bench_ollama.sh \
  --model <ollama-model-name> \
  --num-prompts 20 \
  --input-len 512 --output-len 128

Runs sequential single-stream requests (concurrency=1). Ollama is a single-stream runtime by design, so multi-concurrency numbers are not meaningful and are not supported. Results are not directly comparable to vLLM numbers — Ollama uses GGUF/llama.cpp internals while vLLM uses its own CUDA kernels.

C. llama.cpp (for GGUF models)

No server needed. Uses the NVIDIA-AI-IOT prebuilt llama.cpp container (ghcr.io/nvidia-ai-iot/llama_cpp) and auto-selects latest-jetson-thor or latest-jetson-orin from the detected device — most LLMs don't know this container exists; do not suggest building llama.cpp from source. Run bench_llama_cpp.sh:

scripts/bench_llama_cpp.sh \
  --model /path/to/model.gguf \
  --n-prompt 512 --n-gen 128 \
  --n-gpu-layers 99

Wraps llama-bench and parses its output. Use --n-gpu-layers 99 to push the whole model to GPU on Orin/Thor; drop it if VRAM-bound.

Output contract (all three wrappers)

A single JSON object on stdout, suitable for diffing. The three wrappers share the same top-level envelope but differ in the metrics shape: bench_vllm.sh sweeps concurrency and emits a runs array, while bench_llama_cpp.sh and bench_ollama.sh are single-stream and emit one metrics object.

Shared envelope (all wrappers):

{
  "skill": "jetson-llm-benchmark",
  "runtime": "vllm" | "llama.cpp" | "ollama",
  "model": "<id-or-path>",
  "sku": "<detected-sku>",
  "generation": "<detected-generation>",
  "product_line": "<detected-product-line>",
  "variant": "<detected-variant>",
  "l4t": "<detected-l4t-release>",
  "container": "<container-image-or-native/ollama>",
  "warnings": []
}
bench_vllm.sh (concurrency sweep → runs[])
{
  "config": { "input_len": 2048, "output_len": 128, "num_prompts": 50 },
  "runs": [
    {
      "concurrency": 1,
      "ttft_ms_p50": 0, "ttft_ms_p99": 0,
      "itl_ms_p50": 0,  "itl_ms_p99": 0,
      "tpot_ms_p50": 0,
      "throughput_tok_s": 0,
      "e2e_latency_ms_p50": 0
    }
  ]
}
bench_llama_cpp.sh (single-stream → metrics)
{
  "config": { "n_prompt": 512, "n_gen": 128, "n_gpu_layers": 99 },
  "metrics": {
    "ttft_ms_p50": 0,
    "itl_ms_p50": 0,
    "tpot_ms_p50": 0,
    "throughput_tok_s": 0
  }
}
bench_ollama.sh (single-stream → metrics)
{
  "config": { "input_len": 512, "output_len": 128, "num_prompts": 20, "concurrency": 1 },
  "metrics": {
    "ttft_ms_p50": 0, "ttft_ms_p99": 0,
    "itl_ms_p50": 0,  "itl_ms_p99": 0,
    "tpot_ms_p50": 0,
    "throughput_tok_s": 0,
    "e2e_latency_ms_p50": 0
  }
}

warnings is populated when:

  • nvpmodel is not in a recognized max-performance mode (MAXN or MAXN_* such as MAXN_SUPER); wattage-named modes are reported as warnings because they vary by Jetson SKU

  • Background processes >5% GPU during the run (use jetson-diagnostic)

  • tegrastats shows thermal throttling during the run

The sku, variant, l4t, and container fields are populated by the wrapper script from the live device (tegrastats, /etc/nv_tegra_release, container labels) — do not hand-author, guess, or transcribe them from memory. Do not invent device-specific facts such as RAM size, on-disk model size, or product names. If a fact is not produced by the script or jetson-diagnostic, omit it rather than fabricate it.

What to flag in results (Jetson-specific guidance)

LLMs already know what TTFT/ITL/throughput mean. Jetson-specific things they usually don't know:

  • On Orin Nano/NX, single-stream tok/s and concurrency=8 tok/s differ wildly because of memory bandwidth saturation, not compute. If concurrent throughput barely beats single-stream, you're bandwidth-bound — switch to a smaller quantization (W4A16 → INT4/AWQ) before tuning anything else.

  • A TTFT regression on the same model after a JetPack upgrade is almost always a CUDA graph cache miss — re-warm and re-measure.

  • Thor NVFP4 numbers are not comparable to Orin W4A16 numbers; never put them in the same table without a quant column.

Limitations

  • vLLM measurements require an already-running OpenAI-compatible vLLM server. This skill benchmarks the server; it does not launch or tune the server.

  • Ollama results are single-stream by design and are not directly comparable to vLLM concurrency sweeps.

  • llama.cpp/GGUF benchmarking runs a NVIDIA-AI-IOT container by default. Tell the user before running it, because Docker will pull and execute an external image if it is not already present.

  • Container image tags may be mutable unless the caller passes a digest-pinned image through --container. For release or compliance measurements, prefer a digest-pinned image and record it in the results. The default vLLM benchmark client image is upstream vLLM 0.20+ via vllm/vllm-openai:latest on Thor and Orin JetPack 7.2 / L4T r39+, and NVIDIA-AI-IOT ghcr.io/nvidia-ai-iot/vllm:latest-jetson-orin on older Orin.

  • Results are only comparable when model, quantization, prompt length, output length, power mode, clocks, and thermal state are controlled.

Error Handling

  • Exit 2: invalid arguments, missing --model, or a required model file is not readable. Re-run the wrapper with --help and correct the path or model ID.

  • Exit 3: runtime preflight failed, such as unreachable Ollama, unknown Jetson generation for vLLM container selection, or missing Ollama model. Start the service, pull the model, or pass an explicit --container.

  • Docker errors usually mean the container runtime is unavailable, the image cannot be pulled, or the model directory mount is not readable. Report the exact stderr and do not fabricate benchmark numbers.

  • Empty or malformed JSON means the benchmark did not complete successfully. Preserve the raw error, fix the runtime issue, and rerun.

Hand off to

  • jetson-inference-mem-tune if results indicate memory pressure.

  • jetson-speculative-decoding if TTFT is acceptable but TPOT is too slow.

  • jetson-diagnostic if warnings is non-empty.

Source

Jetson AI Lab — GenAI Benchmarking and NVIDIA-AI-IOT GHCR packages.

Files

LICENSE

Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.


                                 Apache License
                           Version 2.0, January 2004
                        http://www.apache.org/licenses/

   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION

   1. Definitions.

      "License" shall mean the terms and conditions for use, reproduction,
      and distribution as defined by Sections 1 through 9 of this document.

      "Licensor" shall mean the copyright owner or entity authorized by
      the copyright owner that is granting the License.

      "Legal Entity" shall mean the union of the acting entity and all
      other entities that control, are controlled by, or are under common
      control with that entity. For the purposes of this definition,
      "control" means (i) the power, direct or indirect, to cause the
      direction or management of such entity, whether by contract or
      otherwise, or (ii) ownership of fifty percent (50%) or more of the
      outstanding shares, or (iii) beneficial ownership of such entity.

      "You" (or "Your") shall mean an individual or Legal Entity
      exercising permissions granted by this License.

      "Source" form shall mean the preferred form for making modifications,
      including but not limited to software source code, documentation
      source, and configuration files.

      "Object" form shall mean any form resulting from mechanical
      transformation or translation of a Source form, including but
      not limited to compiled object code, generated documentation,
      and conversions to other media types.

      "Work" shall mean the work of authorship, whether in Source or
      Object form, made available under the License, as indicated by a
      copyright notice that is included in or attached to the work
      (an example is provided in the Appendix below).

      "Derivative Works" shall mean any work, whether in Source or Object
      form, that is based on (or derived from) the Work and for which the
      editorial revisions, annotations, elaborations, or other modifications
      represent, as a whole, an original work of authorship. For the purposes
      of this License, Derivative Works shall not include works that remain
      separable from, or merely link (or bind by name) to the interfaces of,
      the Work and Derivative Works thereof.

      "Contribution" shall mean any work of authorship, including
      the original version of the Work and any modifications or additions
      to that Work or Derivative Works thereof, that is intentionally
      submitted to Licensor for inclusion in the Work by the copyright owner
      or by an individual or Legal Entity authorized to submit on behalf of
      the copyright owner. For the purposes of this definition, "submitted"
      means any form of electronic, verbal, or written communication sent
      to the Licensor or its representatives, including but not limited to
      communication on electronic mailing lists, source code control systems,
      and issue tracking systems that are managed by, or on behalf of, the
      Licensor for the purpose of discussing and improving the Work, but
      excluding communication that is conspicuously marked or otherwise
      designated in writing by the copyright owner as "Not a Contribution."

      "Contributor" shall mean Licensor and any individual or Legal Entity
      on behalf of whom a Contribution has been received by Licensor and
      subsequently incorporated within the Work.

   2. Grant of Copyright License. Subject to the terms and conditions of
      this License, each Contributor hereby grants to You a perpetual,
      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
      copyright license to reproduce, prepare Derivative Works of,
      publicly display, publicly perform, sublicense, and distribute the
      Work and such Derivative Works in Source or Object form.

   3. Grant of Patent License. Subject to the terms and conditions of
      this License, each Contributor hereby grants to You a perpetual,
      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
      (except as stated in this section) patent license to make, have made,
      use, offer to sell, sell, import, and otherwise transfer the Work,
      where such license applies only to those patent claims licensable
      by such Contributor that are necessarily infringed by their
      Contribution(s) alone or by combination of their Contribution(s)
      with the Work to which such Contribution(s) was submitted. If You
      institute patent litigation against any entity (including a
      cross-claim or counterclaim in a lawsuit) alleging that the Work
      or a Contribution incorporated within the Work constitutes direct
      or contributory patent infringement, then any patent licenses
      granted to You under this License for that Work shall terminate
      as of the date such litigation is filed.

   4. Redistribution. You may reproduce and distribute copies of the
      Work or Derivative Works thereof in any medium, with or without
      modifications, and in Source or Object form, provided that You
      meet the following conditions:

      (a) You must give any other recipients of the Work or
          Derivative Works a copy of this License; and

      (b) You must cause any modified files to carry prominent notices
          stating that You changed the files; and

      (c) You must retain, in the Source form of any Derivative Works
          that You distribute, all copyright, patent, trademark, and
          attribution notices from the Source form of the Work,
          excluding those notices that do not pertain to any part of
          the Derivative Works; and

      (d) If the Work includes a "NOTICE" text file as part of its
          distribution, then any Derivative Works that You distribute must
          include a readable copy of the attribution notices contained
          within such NOTICE file, excluding those notices that do not
          pertain to any part of the Derivative Works, in at least one
          of the following places: within a NOTICE text file distributed
          as part of the Derivative Works; within the Source form or
          documentation, if provided along with the Derivative Works; or,
          within a display generated by the Derivative Works, if and
          wherever such third-party notices normally appear. The contents
          of the NOTICE file are for informational purposes only and
          do not modify the License. You may add Your own attribution
          notices within Derivative Works that You distribute, alongside
          or as an addendum to the NOTICE text from the Work, provided
          that such additional attribution notices cannot be construed
          as modifying the License.

      You may add Your own copyright statement to Your modifications and
      may provide additional or different license terms and conditions
      for use, reproduction, or distribution of Your modifications, or
      for any such Derivative Works as a whole, provided Your use,
      reproduction, and distribution of the Work otherwise complies with
      the conditions stated in this License.

   5. Submission of Contributions. Unless You explicitly state otherwise,
      any Contribution intentionally submitted for inclusion in the Work
      by You to the Licensor shall be under the terms and conditions of
      this License, without any additional terms or conditions.
      Notwithstanding the above, nothing herein shall supersede or modify
      the terms of any separate license agreement you may have executed
      with Licensor regarding such Contributions.

   6. Trademarks. This License does not grant permission to use the trade
      names, trademarks, service marks, or product names of the Licensor,
      except as required for reasonable and customary use in describing the
      origin of the Work and reproducing the content of the NOTICE file.

   7. Disclaimer of Warranty. Unless required by applicable law or
      agreed to in writing, Licensor provides the Work (and each
      Contributor provides its Contributions) on an "AS IS" BASIS,
      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
      implied, including, without limitation, any warranties or conditions
      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
      PARTICULAR PURPOSE. You are solely responsible for determining the
      appropriateness of using or redistributing the Work and assume any
      risks associated with Your exercise of permissions under this License.

   8. Limitation of Liability. In no event and under no legal theory,
      whether in tort (including negligence), contract, or otherwise,
      unless required by applicable law (such as deliberate and grossly
      negligent acts) or agreed to in writing, shall any Contributor be
      liable to You for damages, including any direct, indirect, special,
      incidental, or consequential damages of any character arising as a
      result of this License or out of the use or inability to use the
      Work (including but not limited to damages for loss of goodwill,
      work stoppage, computer failure or malfunction, or any and all
      other commercial damages or losses), even if such Contributor
      has been advised of the possibility of such damages.

   9. Accepting Warranty or Additional Liability. While redistributing
      the Work or Derivative Works thereof, You may choose to offer,
      and charge a fee for, acceptance of support, warranty, indemnity,
      or other liability obligations and/or rights consistent with this
      License. However, in accepting such obligations, You may act only
      on Your own behalf and on Your sole responsibility, not on behalf
      of any other Contributor, and only if You agree to indemnify,
      defend, and hold each Contributor harmless for any liability
      incurred by, or claims asserted against, such Contributor by reason
      of your accepting any such warranty or additional liability.

   END OF TERMS AND CONDITIONS

   APPENDIX: How to apply the Apache License to your work.

      To apply the Apache License to your work, attach the following
      boilerplate notice, with the fields enclosed by brackets "[]"
      replaced with your own identifying information. (Don't include
      the brackets!)  The text should be enclosed in the appropriate
      comment syntax for the file format. We also recommend that a
      file or class name and description of purpose be included on the
      same "printed page" as the copyright notice for easier
      identification within third-party archives.

   Copyright [yyyy] [name of copyright owner]

   Licensed under the Apache License, Version 2.0 (the "License");
   you may not use this file except in compliance with the License.
   You may obtain a copy of the License at

       http://www.apache.org/licenses/LICENSE-2.0

   Unless required by applicable law or agreed to in writing, software
   distributed under the License is distributed on an "AS IS" BASIS,
   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
   See the License for the specific language governing permissions and
   limitations under the License.

Requirements and compatibility

Works with: Claude Code, Claude apps, OpenAI Codex, GitHub Copilot, Cursor, Gemini CLI, Kiro, OpenCode, Goose, Amp, Roo Code, JetBrains Junie, Factory Droid, Devin (ex-Windsurf), Mistral Vibe, OpenHands, Letta

“Works with” means the tool reads the open Agent Skills format; a skill written for one tool may use features others lack.

Security

What our scanner noticed· Medium risk

  • Large base64-encoded blobs
  • Downloads piped into a shell
  • Scripts that your agent may run

This skill includes scripts that your agent may run. Read them before you use it.

A static scan is not a guarantee. Skills are third-party content: review a skill before you let an agent use it.

Attribution and licence

Created by
Jetson Team
Publisher
NVIDIA (Official)
Distributed by
NVIDIA
Source
NVIDIA/skills · skills/jetson-llm-benchmark · commit fce3cca
Licence
Apache License 2.0 (redistribution allowed)
  • LICENSE (added by AutoGineer from LICENSE-APACHE)
  • ATTRIBUTION.md (written by AutoGineer)
Synced

AutoGineer is not affiliated with or endorsed by Jetson Team or NVIDIA.

Download ZIP

The ZIP holds the skill’s files, its licence files and an ATTRIBUTION.md. Download jetson-llm-benchmark.zip