1 tested weight format · Explore results, then inspect the answers.
Cloud reference · DeepInfra — nvidia/nemotron-3.5-lightning. Quality is scored with the same task bank and graders. All speed and desktop-memory measurements are excluded. Hosted precision and reasoning settings may differ from local GGUF runs.
DeepInfra · Hosted BF16 (provider-reported)
This result tested a cloud endpoint, not a downloadable GGUF.Hosted BF16 (provider-reported) · Not recorded of weights · openrouter · context 16384 · GPU layers unrecorded
Large · 20B and above · MoE · 30.0B total · 3.0B active
Memory fit has not been verified for this configuration.
completed Hosted quality reference; not a local GGUF test. All speed and local memory measurements are excluded.
Temperature 0 · sample index 42 · output limit 4096 · reasoning effort native
RAM labels describe the test computer, not this model’s memory requirement. Speed comparisons keep the 32 GB and 128 GB runs separate.
Runtime build: unrecorded · Cloud · DeepInfra · concurrency: 1
Thinking status: native reasoning · probe cloud-transport-v1 · scope Native reasoning condition; no verified off/on comparison. Provider-reported usage audited against 16K context. Sample indices are not provider seeds. Requested temperature recorded..
Requested off does not prove an absence of reasoning. Saved checks do not establish unrecorded historical template or build details.
69.8 / 100
Nine categories, equally weighted · 9/9 eligible
Complete coverage for this setup. Quality only; speed is separate. How the index works →
Select a category to explore the answers. Bars show what happened on every planned task. Incorrect answers and failures to finish both earn zero; technical problems and untested tasks stay unscored.
Each suite has its own metric and fixed task bank. The balanced index weights categories equally; individual scores and evidence remain available below.
| Test suite | Thinking / Reasoning version | Details |
|---|---|---|
| Everyday coding tasksTask pass rate |
| |
| Larger practical tasksMean task credit |
| |
| Logic and number puzzlesMean puzzle credit |
| |
| Published hard problemsMean answer credit |
| |
| Following exact instructions · original campaignStrict prompt compliance |
| |
| Reading without guessingExact-match accuracy |
| |
| Calculating from a passageNumeric exact-match accuracy |
| |
| Reading research papersAnswer token F1 |
| |
| Resisting incorrect correctionsAll-branches pass rate |
| |
| Following exact instructions · 128 GB campaignStrict prompt compliance | Not planned |
This analysis covers this configuration across its recorded thinking conditions and test suites.
Use this hosted configuration for supervised small project edits, passage calculations, and answers that must track changing source evidence. Nearly all small-coding baseline tasks passed, but larger practical work had low completion credit and research-paper answers were weak. Testing used OpenRouter on DeepInfra at provider-reported BF16 precision; these results do not establish local-hardware performance.
| Tasks passed | 15 of 16 (93.8%) |
|---|---|
| Test coverage | 16 of 16 tasks scored |
| Tasks passed | 7 of 7 (100.0%) |
|---|---|
| Test coverage | 7 of 7 tasks scored |
| Tasks passed | 12 of 12 (100.0%) |
|---|---|
| Test coverage | 12 of 12 tasks scored |
| Average score | 20.7/100 · Answer token F1 |
|---|---|
| Test coverage | 8 of 8 tasks scored |
| Task outcomes | 0 full credit · 5 partial · 3 no credit |
| Average score | 36.4/100 · Mean task credit |
|---|---|
| Test coverage | 11 of 11 tasks scored |
| Task outcomes | 4 full credit · 0 partial · 7 no credit |
Source limitation
Hosted provider/precision condition, not a local GGUF run. Cloud timing and memory are excluded from all local speed and memory comparisons. The hosted output allowance includes reasoning; it does not guarantee the separate local thinking cap. No verified thinking-on/off comparison was run.
Original analysis generated by GPT Astra · · Edited by Codex; agent reviewed by Codex · publication authorized by user
Codex reviewed this assessment against a newer evidence edition. The original analyst, generation date and source draft remain recorded separately from the editorial revision.
Generation and editorial authorship remain attached to this revision. Changing the default generator does not change this attribution.
Handles passage calculations and keeps answers aligned with source evidence, resisting incorrect corrections while updating when the source changes. It passed every test in these small sets—not proof of broad expertise.
| Tasks passed | 7 of 7 (100.0%) |
|---|---|
| Test coverage | 7 of 7 tasks scored |
| Tasks passed | 12 of 12 (100.0%) |
|---|---|
| Test coverage | 12 of 12 tasks scored |
Finding information in small projects, editing files and fixing bugs are useful applications: nearly all baseline tasks passed. Repeated attempts also often succeeded, but an individual success does not guarantee dependable automation.
| Tasks passed | 15 of 16 (93.8%) |
|---|---|
| Test coverage | 16 of 16 tasks scored |
| Repetitions earning full credit | 47 repetitions |
|---|---|
| Repetitions scored | 48 repetitions |
| Repetitions planned | 48 repetitions |
| Questions with all repetitions scored | 16 questions |
| Distinct questions planned | 16 questions |
Research-paper answers were poor across the complete test, with low answer-matching scores—not merely citation-format mistakes. Do not rely on it for paper-based answers without checking the original text.
| Average score | 20.7/100 · Answer token F1 |
|---|---|
| Test coverage | 8 of 8 tasks scored |
| Task outcomes | 0 full credit · 5 partial · 3 no credit |
Larger projects and difficult reasoning need close checking. Larger practical tasks earned low completion credit, while puzzle and competition-math results were mixed. The totals include confirmed failures to finish, which should not be mistaken for incorrect submitted answers.
| Average score | 36.4/100 · Mean task credit |
|---|---|
| Test coverage | 11 of 11 tasks scored |
| Task outcomes | 4 full credit · 0 partial · 7 no credit |
| Average score | 60.5/100 · Mean puzzle credit |
|---|---|
| Test coverage | 34 of 34 tasks scored |
| Task outcomes | 20 full credit · 1 partial · 13 no credit |
| Average score | 47.1/100 · Mean answer credit |
|---|---|
| Test coverage | 11 of 11 tasks scored |
| Task outcomes | 5 full credit · 2 partial · 4 no credit |
Confirmed failures to finish within the test budget earn zero completion credit and are shown separately from incorrect submitted answers. Saved work can earn credit without a final chat message. Technical failures and uncertain answer-channel failures remain unscored; authorized infrastructure retries count once per scheduled trial.
Source limitation
Full and partial credit follow saved task checks. Incorrect answers and confirmed failures to finish within the test budget both earn zero, but are shown separately. Provider failures, ambiguous answer-channel failures and uncertain context guards remain unscored. A task blocked before execution is untested. Saved work can earn credit without a final chat message. Cloud timing is excluded. Authorized infrastructure retries count once per scheduled trial; original failures remain in history.
This hosted condition cannot establish local speed or memory requirements. Reasoning shares the output allowance. Native reasoning was tested without a verified on/off comparison, so its benefit and a toggle recommendation cannot be established.
Source limitation
Hosted provider/precision condition, not a local GGUF run. Cloud timing and memory are excluded from all local speed and memory comparisons. The hosted output allowance includes reasoning; it does not guarantee the separate local thinking cap. No verified thinking-on/off comparison was run.
Publication redactions and source-link-only paper material limit readers’ inspection of some examples. These restrictions do not establish that the model lacked the paper.
Source limitation
Some selected evidence has recorded publication redactions.
One weight format is recorded for this model. Additional tested formats will appear here when available.
No verified on/off comparison is available for this configuration and test suite. Requested modes alone do not establish a controlled thinking comparison.
Repeated runs are a separate experiment and never add votes to the baseline score.
48/48 repeats scored · 47 full-credit repeats.
16/16 questions have every planned repeat scored. Of those, 15 earned full credit on every repeat.
Find information in a small project, edit files and fix bugs.
Each segment counts tasks, not credit points. Partial credit varies by task.
Scoring and fixed question bank →
Select a question to read the original answer, tool calls and available code changes for this configuration and thinking setting.
1–16 of 16 matching questions · 16 in this version
The user asks where the project's retry-limit setting is defined and what value it holds. This exercises locating a named setting in code and reporting both its location and value.
basic-T1 · primaryThe user asks for the release date mentioned in a project notes document. This exercises reading a specified file and extracting a particular fact rather than guessing from outside knowledge.
basic-T2 · primaryThe user asks to change the project's retry limit to a specified value. This exercises finding the relevant setting and making a focused code edit rather than merely describing the change.
basic-T3 · primaryThe user asks to create a text file containing exactly one supplied line. This exercises basic file creation and precise instruction following, including avoiding extra text in the file.
basic-T4 · primaryThe user asks what the most recent Git commit changed in the project. This exercises inspecting recorded changes and turning them into a clear explanation of what was modified.
basic-T5 · primaryThe user asks how many Python files are in the application folder and which is largest. This exercises identifying relevant files, counting them, and comparing their sizes.
basic-T6 · primaryThe user asks to add a function that converts byte counts into readable sizes and remove the related TODO comment. This exercises implementing a small utility and updating the surrounding code consistently.
basic-T7 · primaryThe user asks to find every TODO comment in the Python code and append each to the project's task list on a separate line. This combines searching across files with collecting text into an existing document.
basic-T8 · primaryThe user asks to investigate a failing test and fix the application code without changing the test. This exercises diagnosing a bug while respecting an explicit limit on what may be edited.
basic-H1 · primaryThe user asks to rename a network function and update every place in the project that uses it. This exercises tracking references across files and making a coordinated change rather than editing only the definition.
basic-H2 · primaryThe user asks to find a proxy configuration file somewhere under the project's parent folder and report its host and port. This exercises searching beyond the project folder and reading configuration values.
basic-H3 · primaryThe user asks which error code appears in a long event log and on which line. This exercises finding a specific entry in a large file and reporting its precise location.
basic-H4 · primaryThe user asks to append a supplied release-related task as the final line of an existing task list. This exercises preserving existing content while adding exact text in the requested position.
basic-H5 · primaryThe user asks for the total time a network function spends sleeping when all attempts fail, worked out from the code. This exercises following retry logic and adding the delays along that failure path.
basic-H6 · primaryThe user asks for a project readme listing every command-line option with its help text. This exercises discovering the program's available options and transferring that information into complete, source-grounded documentation.
basic-H7 · primaryThe user asks to undo the latest Git commit's change by editing the file directly, without using Git to make changes. This exercises understanding a previous edit and reversing it under a method constraint.
basic-H8 · primaryThese are starting points for inspection, not claims that one setup is better. Runtime and placement differences remain visible.
Another checkpoint in the same total-parameter band and architecture category, with matching task controls.
Requested thinking on · Complete compared coverageCompare these configurations →
DeepInfra · recorded source
Choose a suite to inspect its exact available questions, references and model answers.