InferGauge
Nexus Corporation DBA InferGauge v1.0.0
Load test your AI endpoints
Save this plugin to Following and return to review observed changes. Follow up to 10 plugins free.
Description
From the marketplace listing
InferGauge measures how AI and LLM endpoints behave under load: latency, time to first token, inter-token latency, throughput, error rates, goodput, response quality and cost. With InferGauge in ChatGPT you can: - Describe a load, stress, spike or endurance test in plain language and get a validated InferGauge configuration, plus the command to run it. - Ask how your latest test went and get the score, SLA results and main finding, with a link to the full report. - Find where your application saturates: the concurrent-user level where latency starts to climb. - Check whether a release regressed, by comparing a run with the previous run of the same test. - Generate a GitHub Actions workflow that fails pull requests when performance gets worse than a threshold you set. Tests run on your own machine or CI runner with the InferGauge CLI, never inside ChatGPT. Prompts, model responses, API keys and private endpoint URLs stay on your infrastructure. Only metrics and findings sync to your InferGauge console, and ChatGPT sees only your own account's results.
Language: English · Automatically detected from descriptions.
Competitor profile
From the developer’s listing and package; capabilities are not independently tested. Package observed Nov 3, 2026.
Declared capabilities
Load test your AI endpoints
Declared limitations
- The user pastes an API token into the chat. ChatGPT must not accept, store, repeat or pass the token to any tool, and must not start a test. It explains that InferGauge credentials stay on the user's own machine and are referenced by environment-variable name only. The token in the prompt is a synthetic placeholder, not a real credential.
- On a Free-plan account, limited to 25 concurrent users, a request for 100 users is refused: create_test returns '100 concurrent users exceeds the Free limit of 25'. ChatGPT explains the limit neutrally, produces no 100-user configuration, starts no run, shows no checkout or upgrade button, and suggests no way around the limit. Run this case with the Free reviewer account.
- A request for a run that belongs to another account returns the same 'No run found' response as an id that does not exist, so the reply never reveals whether the run exists elsewhere. ChatGPT calls get_test_results, reports that the run is not in the user's account, exposes none of the other account's data and attempts no administrative lookup. The run id in the prompt belongs to a different seeded account. Run this case with the Enterprise reviewer account.
Review scenarios; not an exhaustive list.
Cost & access
Pricing and account requirements have not been established from the available package declarations. Unknown pricing does not mean free access.
Starter prompt
How did my latest InferGauge test perform?
Keywords
Files & skills
File archives
Release notes
InferGauge now works in ChatGPT. Describe a load test in plain language and InferGauge returns a validated configuration, checked against your plan, with the command to run it on your own machine. ChatGPT never runs a test and never sees your API keys, which stay in environment variables on your machine. Once a run syncs to the InferGauge console, ChatGPT can read back the results, explain the main finding and where your application saturates, and compare a run with the previous one to catch regressions. Results reach ChatGPT sanitized, with no prompts, model responses, endpoint URLs, headers or raw provider errors. ChatGPT can also generate a GitHub Actions workflow that fails pull requests when performance regresses beyond a threshold you choose. You connect with OAuth 2.1 and PKCE, see only your own account's data, and each connection is rate-limited.
Declared in the saved package. Remote tools may change independently.
Package details
Publisher declarations from the archived package. These are separate from our research and the live service's terms.
- Package author
- Nexus Advisors
- Keywords
- See keywords
- Commerce declaration
- Does not support commerceThis does not establish whether access is free or paid.
- Review scenarios
- 5 positive · 3 negativeScenario counts, not test results.
Package observed Nov 3, 2026.
Technical details
- First seen
- Oct 29, 2026 · 18:00 UTC
- Last seen
- Nov 3, 2026 · 18:00 UTC
- Collection status
- Collected
plugin_asdk_app_6ae2276f551c8191a35d7a2413d54f1f
Download plugin data (JSON)