AgentBench / results
← All models

Gemma 4 31B · CoreWeave (cloud)

1 tested weight format · Explore results, then inspect the answers.

Cloud reference · CoreWeave — google/gemma-4-31b-it. 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.

Choose a tested configuration

Hosted model source

CoreWeave · Hosted FP4 (provider-reported)

This result tested a cloud endpoint, not a downloadable GGUF.
Open hosted model →

Setup, memory & provenance · Hosted FP4 (provider-reported)

Hosted FP4 (provider-reported) · Not recorded of weights · openrouter · context 16384 · GPU layers unrecorded

Large · 20B and above · dense · 30.7B total

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.

Recorded settings and thinking verification

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 · CoreWeave · 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.

Overall across tasks

73.0 / 100

Nine categories, equally weighted · 9/9 eligible

Complete coverage for this setup. Quality only; speed is separate. How the index works →

What does it do well?

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.

Exact scores and task times by suite

Each suite has its own metric and fixed task bank. The balanced index weights categories equally; individual scores and evidence remain available below.

Test suiteThinking / Reasoning versionDetails
Everyday coding tasksTask pass rate
Pass rate
16/16 · 100%
Answers completed
16/16 · 100%
Correct / completed
16/16 · 100%
16 full credit
16/16 scoredNo measured time
Larger practical tasksMean task credit
Average credit
63.6%
Full-credit rate
7/11 · 64%
Answers completed
8/11 · 73%
Full credit / completed
7/8 · 88%
7 full credit1 incorrect3 failed to finish
11/11 scoredNo measured time
Logic and number puzzlesMean puzzle credit
Average credit
20.6%
Full-credit rate
7/34 · 21%
Answers completed
10/34 · 29%
Full credit / completed
7/10 · 70%
7 full credit3 incorrect24 failed to finish
34/34 scoredNo measured time
Published hard problemsMean answer credit
Average credit
54.5%
Full-credit rate
6/11 · 55%
Answers completed
6/11 · 55%
Full credit / completed
6/6 · 100%
6 full credit5 failed to finish
11/11 scoredNo measured time
Following exact instructions · original campaignStrict prompt compliance
Pass rate
17/23 · 74%
Answers completed
19/23 · 83%
Correct / completed
17/19 · 89%
17 full credit2 incorrect4 failed to finish
23/23 scoredNo measured time
Reading without guessingExact-match accuracy
Pass rate
16/16 · 100%
Answers completed
16/16 · 100%
Correct / completed
16/16 · 100%
16 full credit
16/16 scoredNo measured time
Calculating from a passageNumeric exact-match accuracy
Pass rate
7/7 · 100%
Answers completed
7/7 · 100%
Correct / completed
7/7 · 100%
7 full credit
7/7 scoredNo measured time
Reading research papersAnswer token F1
Average credit
44.3%
Full-credit rate
0/8 · 0%
Answers completed
8/8 · 100%
Full credit / completed
0/8 · 0%
8 partial credit
8/8 scoredNo measured time
Resisting incorrect correctionsAll-branches pass rate
Pass rate
12/12 · 100%
Answers completed
12/12 · 100%
Correct / completed
12/12 · 100%
12 full credit
12/12 scoredNo measured time
Following exact instructions · 128 GB campaignStrict prompt complianceNot planned

Model analysis

This analysis covers this configuration across its recorded thinking conditions and test suites.

Use this hosted native-reasoning model for small coding jobs and answering questions from passages. It also succeeded on most larger practical tasks, making it worth trying beyond simple edits—but check that the work finishes correctly. Difficult logic and number puzzles are a poor fit: most attempts did not finish within the test budget. These results support a scoped assistant, not an unattended problem-solver, and do not establish local performance.

Evidence for this judgment
Everyday coding tasks · Thinking / Reasoning version
Tasks passed16 of 16 (100.0%)
Test coverage16 of 16 tasks scored
Explore these test results →

View this model’s answers, code and tool activity →

Reading without guessing · Thinking / Reasoning version
Tasks passed16 of 16 (100.0%)
Test coverage16 of 16 tasks scored
Explore these test results →

View this model’s answers, code and tool activity →

Larger practical tasks · Thinking / Reasoning version
Average score63.6/100 · Mean task credit
Test coverage11 of 11 tasks scored
Task outcomes7 full credit · 0 partial · 4 no credit
Failed to finish3 tasks
Incorrect completed answers1 tasks
Explore these test results →

View this model’s answers, code and tool activity →

Logic and number puzzles · Thinking / Reasoning version
Average score20.6/100 · Mean puzzle credit
Test coverage34 of 34 tasks scored
Task outcomes7 full credit · 0 partial · 27 no credit
Failed to finish24 tasks
Explore these test results →

View this model’s answers, code and tool activity →

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.

Summary generated by GPT Astra · · Agent reviewed by Codex · publication authorized by user

About this summary
Analyst provider
openai
Generation route
Codex CLI · codex-cli 0.163.0 · ChatGPT account
Requested model
gpt-6-astra
Returned model
Not supplied; identity is recorded from the request only.
Exact provider snapshot
Exact provider snapshot not supplied.
Reasoning effort
high
Prompt version
model-summary-reader-v8
Generation time
2026-10-22T06:19:22.097520Z
Review time
2026-10-22T06:50:55.237684+00:00
Evidence reviewer
Codex (agent)
Publication authorization
User · 2026-10-22T05:55:43.625712+00:00 · Go ahead and complete the remaining work. Spin up agents: luna, terra, sol, astra, as needed to get it done. Narrative summaries: these are the astra generated summaries? Go ahead and get those completed as well. Once all of this is done, you may push the completed result.
Originating evidence edition
dataset-5a6f96ec56c3d240cc42fa23147c1495da1be71f3ef0722c0e5f8c4f5e126a83
Analysis input hash
c62aab2e7d9099be361da4de307fd29314101a76789e9311b41663ee3e9a4cfb
Summary revision
summary-df9ea23ae8791adb6c5df9ab4e315f4fcf4f7b35c0d36cbe866ac906a604ed35

Generation and editorial authorship remain attached to this revision. Changing the default generator does not change this attribution.

Good at

Handles small project edits and bug fixes well on these tests. Success extended across almost all sampled repetitions, supporting everyday coding use without establishing the same consistency on larger projects.

Evidence for this judgment
Everyday coding tasks · Thinking / Reasoning version
Tasks passed16 of 16 (100.0%)
Test coverage16 of 16 tasks scored
Explore these test results →

View this model’s answers, code and tool activity →

Everyday coding tasks · Thinking on · sampled repetitions
Repetitions earning full credit47 repetitions
Repetitions scored48 repetitions
Repetitions planned48 repetitions
Questions with all repetitions scored16 questions
Distinct questions planned16 questions
Explore these test results →

View this model’s answers, code and tool activity →

Answers passage questions and calculations accurately, including recognizing missing answers. It passed all these small tests and the source-backed correction tests: maintaining answers against incorrect challenges while updating them when the source changed.

Evidence for this judgment
Calculating from a passage · Thinking / Reasoning version
Tasks passed7 of 7 (100.0%)
Test coverage7 of 7 tasks scored
Explore these test results →

View this model’s answers, code and tool activity →

Reading without guessing · Thinking / Reasoning version
Tasks passed16 of 16 (100.0%)
Test coverage16 of 16 tasks scored
Explore these test results →

View this model’s answers, code and tool activity →

Resisting incorrect corrections · Thinking / Reasoning version
Tasks passed12 of 12 (100.0%)
Test coverage12 of 12 tasks scored
Explore these test results →

View this model’s answers, code and tool activity →

Struggles with

Struggles to finish demanding reasoning work. Most logic-and-number puzzles ended unfinished; difficult published math and reasoning questions also had substantial noncompletion, rather than merely wrong final answers.

Evidence for this judgment
Logic and number puzzles · Thinking / Reasoning version
Average score20.6/100 · Mean puzzle credit
Test coverage34 of 34 tasks scored
Task outcomes7 full credit · 0 partial · 27 no credit
Failed to finish24 tasks
Incorrect completed answers3 tasks
Explore these test results →

View this model’s answers, code and tool activity →

Published hard problems · Thinking / Reasoning version
Average score54.5/100 · Mean answer credit
Test coverage11 of 11 tasks scored
Task outcomes6 full credit · 0 partial · 5 no credit
Failed to finish5 tasks
Incorrect completed answers0 tasks
Explore these test results →

View this model’s answers, code and tool activity →

Research-paper answers need careful checking: answer matching was modest, with partial matches throughout. Exact wording and formatting are another boundary; some instruction-following tasks were incorrect, while others did not finish.

Evidence for this judgment
Reading research papers · Thinking / Reasoning version
Average score44.3/100 · Answer token F1
Test coverage8 of 8 tasks scored
Task outcomes0 full credit · 8 partial · 0 no credit
Partially correct answers8 tasks
Explore these test results →

View this model’s answers, code and tool activity →

Following exact instructions · Thinking / Reasoning version
Tasks passed17 of 23 (73.9%)
Test coverage23 of 23 tasks scored
Incorrect completed answers2 tasks
Failed to finish4 tasks
Explore these test results →

View this model’s answers, code and tool activity →

Worth knowing

This hosted test does not establish local speed or memory requirements. Only native reasoning was tested, so changing thinking settings cannot be recommended. The hosted output allowance includes reasoning, unlike a separate local thinking allowance.

Evidence for this judgment

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.

Wrong answers and failures to finish are distinguished; uncertain infrastructure outcomes remain unassessed. Authorized retries count once, and completed work can qualify without a final chat message.

Evidence for this judgment

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.

Publication redactions limit reader inspection of some evidence.

Evidence for this judgment

Source limitation
Some selected evidence has recorded publication redactions.

How to interpret these assessments and their limitations →

One weight format is recorded for this model. Additional tested formats will appear here when available.

Thinking tradeoff

No verified on/off comparison is available for this configuration and test suite. Requested modes alone do not establish a controlled thinking comparison.

Sampled repetitions

Repeated runs are a separate experiment and never add votes to the baseline score.

Thinking / Reasoning version · temperature 0.6

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.

Recorded repetition conditions
  • Seed 303 · requested on · native reasoning
  • Seed 202 · requested on · native reasoning
  • Seed 101 · requested on · native reasoning

Everyday coding tasks: recorded outcomes

Find information in a small project, edit files and fix bugs.

16 full credit

Each segment counts tasks, not credit points. Partial credit varies by task.

Scoring and fixed question bank →

This model’s answers, code and tool activity

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

  • Locate the Retry Limit

    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 · primary
  • Find the Release Date

    The 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 · primary
  • Change the Retry Limit

    The 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 · primary
  • Create an Exact One-Line File

    The 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 · primary
  • Explain the Latest Commit

    The 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 · primary
  • Count and Compare Python Files

    The 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 · primary
  • Add Readable Byte Sizes

    The 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 · primary
  • Collect TODO Comments

    The 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 · primary
  • Fix Code Behind a Failing Test

    The 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 · primary
  • Rename a Function Throughout

    The 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 · primary
  • Find the Proxy Settings

    The 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 · primary
  • Locate an Error in a Log

    The 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 · primary
  • Append a Final Task

    The 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 · primary
  • Calculate Total Retry Waiting

    The 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 · primary
  • Document Every Command-Line Option

    The 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 · primary
  • Undo a Commit by Editing

    The 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 · primary
Filters change this list, not the fixed scoring denominator. Search currently covers question IDs, titles and available task explanations. Focus a control to load prompt/outcome search where this edition provides it.

Suggested comparisons

These are starting points for inspection, not claims that one setup is better. Runtime and placement differences remain visible.

Gemma 4 31B · Cloud / Novita

Another checkpoint in the same total-parameter band and architecture category, with matching task controls.

Requested thinking on · Complete compared coverage

Compare these configurations →

Selection rule and evidence IDsPolicy comparisons-v3 · Template same-parameter-architecture-matching-controls
Facts: fact-2479d33a423c279b90dac82f, fact-9e913670a7bd275066eb0666
Matching candidate set: fact-9e913670a7bd275066eb0666

Tested weights

Hosted FP4 (provider-reported) · Not recorded

CoreWeave · recorded source

Explore the answers

Choose a suite to inspect its exact available questions, references and model answers.