AgentBench / results
← All models

Gemma 4 26B-A4B

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

Choose a tested configuration

Get this tested model

UD-Q4_K_M · 16.95 GB download

Links follow the selected quant and configuration.

Setup, memory & provenance · UD-Q4_K_M

UD-Q4_K_M · 16.95 GB of weights · rocm · context 32768 · GPU layers 99

Large · 20B and above · MoE · 25.2B total · Active parameter count not recorded

Memory fit has not been verified for this configuration.

completed

Recorded settings and thinking verification

Temperature 0 · seed 42 · output limit 4096 · reasoning effort unrecorded

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 · Local · test machine with 32 GB RAM · concurrency: 1

Thinking status: verified · probe 1 · scope Saved profile-level probe; exact historical template/build scope is not fully recorded..

Requested off does not prove an absence of reasoning. Saved checks do not establish unrecorded historical template or build details.

Overall across tasks

80.3 / 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 offDetails
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 scored10.6s median task time · 16 measured
Larger practical tasksMean task credit
Average credit
90.9%
Full-credit rate
10/11 · 91%
Answers completed
11/11 · 100%
Full credit / completed
10/11 · 91%
10 full credit1 incorrect
11/11 scored54.7s median task time · 11 measured
Logic and number puzzlesMean puzzle credit
Average credit
82.4%
Full-credit rate
28/34 · 82%
Answers completed
34/34 · 100%
Full credit / completed
28/34 · 82%
28 full credit1 partial credit5 incorrect
34/34 scored62.2s median task time · 34 measured
Published hard problemsMean answer credit
Average credit
69.1%
Full-credit rate
7/11 · 64%
Answers completed
11/11 · 100%
Full credit / completed
7/11 · 64%
7 full credit1 partial credit3 incorrect
11/11 scored53.6s median task time · 11 measured
Following exact instructions · original campaignStrict prompt compliance
Pass rate
19/23 · 83%
Answers completed
23/23 · 100%
Correct / completed
19/23 · 83%
19 full credit4 incorrect
23/23 scored16.9s median task time · 23 measured
Reading without guessingExact-match accuracy
Pass rate
14/16 · 88%
Answers completed
16/16 · 100%
Correct / completed
14/16 · 88%
14 full credit2 incorrect
16/16 scored1.2s median task time · 16 measured
Calculating from a passageNumeric exact-match accuracy
Pass rate
6/7 · 86%
Answers completed
7/7 · 100%
Correct / completed
6/7 · 86%
6 full credit1 incorrect
7/7 scored1.5s median task time · 7 measured
Reading research papersAnswer token F1
Average credit
24.8%
Full-credit rate
0/8 · 0%
Answers completed
8/8 · 100%
Full credit / completed
0/8 · 0%
4 partial credit4 incorrect
8/8 scored8.7s median task time · 8 measured
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 scored2.9s median task time · 12 measured
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 tested Gemma configuration for small project edits and longer practical tasks with thinking off. It passed all the small coding tasks and completed most larger tasks, so its usefulness extends beyond toy edits. Small coding tasks had a median total time of 10.6 seconds on this test machine—a useful balance of responsiveness and successful work. Research-paper answering was poor, however; extracting dependable answers from papers is a clear boundary.

Evidence for this judgment
Everyday coding tasks · Thinking off
Tasks passed16 of 16 (100.0%)
Test coverage16 of 16 tasks scored
Tasks timed16 tasks
Typical time per task (median)10.6 seconds
Explore these test results →

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

Larger practical tasks · Thinking off
Average score90.9/100 · Mean task credit
Test coverage11 of 11 tasks scored
Task outcomes10 full credit · 0 partial · 1 no credit
Explore these test results →

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

Reading research papers · Thinking off
Average score24.8/100 · Answer token F1
Test coverage8 of 8 tasks scored
Task outcomes0 full credit · 4 partial · 4 no credit
Explore these test results →

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

Original analysis generated by GPT Astra · · Edited by Codex; agent reviewed by Codex · publication authorized by user

About this summary

Codex reviewed this assessment against a newer evidence edition. The original analyst, generation date and source draft remain recorded separately from the editorial revision.

Analyst provider
openai
Generation route
Codex CLI · codex-cli 0.162.1 · 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-v6
Generation time
2026-10-12T05:13:56.630216Z
Review time
2026-10-22T05:55:43.646587+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-b1aa28e61ea9ce402d5617e636622c22565093f8849fd66537b5f4353d228521
Analysis input hash
cc0ef91c51d9139fff921fe48f072c7d1b3bd98a6cde419d5c4e3e5e7cacb82d
Original analysis input hash
d6ba23618bfd38086b39cf3f8832a25ed089dce9ac662b56f4001a00268b0f41
Editorial evidence edition
dataset-5a6f96ec56c3d240cc42fa23147c1495da1be71f3ef0722c0e5f8c4f5e126a83
Original summary
summary-30dc8a92b50ecd356171c8138fb4c6618ed15eb58c81ea94c4954c5c5b6b84e4
Summary revision
summary-082b39361c5a388021b81e3fc81efd8b7ef737d18ea4f698f75a76289bbb8443

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

Contributing analyses (22)
  • GPT Astra ·
    Analysis stage
    evidence_segment
    Requested / returned model
    gpt-6-astra / Not supplied
    Exact provider snapshot
    Exact provider snapshot not supplied.
    Reasoning effort
    high
    Prompt version
    model-summary-v3
    Analysis input hash
    1cc3f43bf9aa4fa653dd6da99286f0ba2853d9a9e8e26fe1ca0952aa20ba5a5e
    Summary revision
    summary-072b62a90f03a8a897d9c1e5bec2c457dc1342b4b264fc2eb1abf1ccae8da10d
    Input summary revisions
    Direct evidence analysis
  • GPT Astra ·
    Analysis stage
    evidence_segment
    Requested / returned model
    gpt-6-astra / Not supplied
    Exact provider snapshot
    Exact provider snapshot not supplied.
    Reasoning effort
    high
    Prompt version
    model-summary-v3
    Analysis input hash
    efdf29e8d62c8d2efa6ff175928f237e59a7253e38ff2c67f7252bca9e369c9b
    Summary revision
    summary-0732052a3b0dd13cba04da2aad82f18250d5c966fd1c029887ee76029a4dde9c
    Input summary revisions
    Direct evidence analysis
  • GPT Astra ·
    Analysis stage
    suite_analysis
    Requested / returned model
    gpt-6-astra / Not supplied
    Exact provider snapshot
    Exact provider snapshot not supplied.
    Reasoning effort
    high
    Prompt version
    model-summary-v3
    Analysis input hash
    385362754b3cc6e0b481dbe0ba2bc53d4f9e3e2046ca752578dbeaa58b137b0a
    Summary revision
    summary-1254de06c11ac61e31548852db15b8d27ce37b047b15bfeeb43668be69256dbe
    Input summary revisions
    Direct evidence analysis
  • GPT Astra ·
    Analysis stage
    evidence_segment
    Requested / returned model
    gpt-6-astra / Not supplied
    Exact provider snapshot
    Exact provider snapshot not supplied.
    Reasoning effort
    high
    Prompt version
    model-summary-v3
    Analysis input hash
    86186b455ca89eef604cde354194cd45423a2ed6ca5ba6fd210afa851b7bf23f
    Summary revision
    summary-1962ce22c342d6b7181c7524b56f9cae2d508cb3cfddea5931ea4e1d3c67e800
    Input summary revisions
    Direct evidence analysis
  • GPT Astra ·
    Analysis stage
    suite_analysis
    Requested / returned model
    gpt-6-astra / Not supplied
    Exact provider snapshot
    Exact provider snapshot not supplied.
    Reasoning effort
    high
    Prompt version
    model-summary-v3
    Analysis input hash
    cd81ae44d6cbc85c7a61a0554b366ebc5080902ebcbb78c0c9e14d74b33b424c
    Summary revision
    summary-1e64d3a82c390ea59249dec9cc93679cdfb9117191c6120a04baf4341630cd9a
    Input summary revisions
    Direct evidence analysis
  • GPT Astra ·
    Analysis stage
    suite_analysis
    Requested / returned model
    gpt-6-astra / Not supplied
    Exact provider snapshot
    Exact provider snapshot not supplied.
    Reasoning effort
    high
    Prompt version
    model-summary-v3
    Analysis input hash
    fd992d9f04494500c0696e5ded6a8771c61a5e0923bfc3e4921ecc2488f36bea
    Summary revision
    summary-289862be9dbc50fed4097c9db33fdf0d4530b2592cecc51cf9371aa8dd939e62
    Input summary revisions
    Direct evidence analysis
  • GPT Astra ·
    Analysis stage
    suite_analysis
    Requested / returned model
    gpt-6-astra / Not supplied
    Exact provider snapshot
    Exact provider snapshot not supplied.
    Reasoning effort
    high
    Prompt version
    model-summary-v3
    Analysis input hash
    d8c9c422a46bc30b58a59ee0d7416a656f3b180b88d0203abb97a102b030559e
    Summary revision
    summary-3e6f27b46e15ff354e30d6b9bc21c10f68d629ea1aec2a66aad2d4a3fb3461e5
    Input summary revisions
    Direct evidence analysis
  • GPT Astra ·
    Analysis stage
    suite_analysis
    Requested / returned model
    gpt-6-astra / Not supplied
    Exact provider snapshot
    Exact provider snapshot not supplied.
    Reasoning effort
    high
    Prompt version
    model-summary-v3
    Analysis input hash
    5a167a6a7fea349aec57f47d4eca1326b44075e078a2903281913bea74211739
    Summary revision
    summary-49fd0c0ec45052b1698170803cc6f31f0bb2482887f500ba0342c6a47b8556eb
    Input summary revisions
    Direct evidence analysis
  • GPT Astra ·
    Analysis stage
    example_analysis
    Requested / returned model
    gpt-6-astra / Not supplied
    Exact provider snapshot
    Exact provider snapshot not supplied.
    Reasoning effort
    high
    Prompt version
    model-summary-v3
    Analysis input hash
    8334554eca79b78f468441eb74fc9790ded2889ad652b30b935c6014d77f5e40
    Summary revision
    summary-60fc6c2721b30bf1b93fce9cacfb79f0aae17b1b608af27364d92669273fb2d9
    Input summary revisions
    Direct evidence analysis
  • GPT Astra ·
    Analysis stage
    subject_synthesis
    Requested / returned model
    gpt-6-astra / Not supplied
    Exact provider snapshot
    Exact provider snapshot not supplied.
    Reasoning effort
    high
    Prompt version
    model-summary-v3
    Analysis input hash
    240da9a85ff1e5716640d0119b04b859d5d0affe19f048d7e9ab0b0395f95dfb
    Summary revision
    summary-752cb0e0f166c85f4e308810735cb5979c50691f45ca4b68b2747e0d361d7514
    Input summary revisions
    summary-1254de06c11ac61e31548852db15b8d27ce37b047b15bfeeb43668be69256dbe, summary-1e64d3a82c390ea59249dec9cc93679cdfb9117191c6120a04baf4341630cd9a, summary-289862be9dbc50fed4097c9db33fdf0d4530b2592cecc51cf9371aa8dd939e62, summary-3e6f27b46e15ff354e30d6b9bc21c10f68d629ea1aec2a66aad2d4a3fb3461e5, summary-49fd0c0ec45052b1698170803cc6f31f0bb2482887f500ba0342c6a47b8556eb, summary-76b8e8f18ec0985f884a9083351b6cdf45075617fc401a3eb4c56b2ede63c2f6, summary-aee90cba6d16163810ae20b3705c1f61cdbb842407337a92c4e799f1a1541aa0, summary-c6e38cfb1f4af7f348f665549dab55bd2afee1219e0a1ee9f686f20ec7403665, summary-deba09ea9bb8095260bd593b6b83a7d0fe0716e15da2a864cdad6c9f8ba1bc87
  • GPT Astra ·
    Analysis stage
    suite_analysis
    Requested / returned model
    gpt-6-astra / Not supplied
    Exact provider snapshot
    Exact provider snapshot not supplied.
    Reasoning effort
    high
    Prompt version
    model-summary-v3
    Analysis input hash
    0f31ddac8f162f9a33637c490284ff41e8d23d76910d29bc888490f22978ea22
    Summary revision
    summary-76b8e8f18ec0985f884a9083351b6cdf45075617fc401a3eb4c56b2ede63c2f6
    Input summary revisions
    Direct evidence analysis
  • GPT Astra ·
    Analysis stage
    example_analysis
    Requested / returned model
    gpt-6-astra / Not supplied
    Exact provider snapshot
    Exact provider snapshot not supplied.
    Reasoning effort
    high
    Prompt version
    model-summary-v3
    Analysis input hash
    5171859c7fb6e6d52d65cb717abdc61c515c9e8133db9da13f888369a92e5cf8
    Summary revision
    summary-76d5e82b6a8424978f0074829d5d9d2f879d3baabc5c993f30b811c88f38621c
    Input summary revisions
    Direct evidence analysis
  • GPT Astra ·
    Analysis stage
    evidence_segment
    Requested / returned model
    gpt-6-astra / Not supplied
    Exact provider snapshot
    Exact provider snapshot not supplied.
    Reasoning effort
    high
    Prompt version
    model-summary-v3
    Analysis input hash
    fa4928bcacfdaeb44664092b78297e0677c85d0d72415b6f5f97f54fa800bbb8
    Summary revision
    summary-8030273b38146b827f397bd46e34cc296edb00ca2be4f3d68e781e01d5f5a101
    Input summary revisions
    Direct evidence analysis
  • GPT Astra ·
    Analysis stage
    evidence_segment
    Requested / returned model
    gpt-6-astra / Not supplied
    Exact provider snapshot
    Exact provider snapshot not supplied.
    Reasoning effort
    high
    Prompt version
    model-summary-v3
    Analysis input hash
    1c5bd29a6da45a00658ba9a42831bf64ed43ad8f98995833175c263086171335
    Summary revision
    summary-9b2c30abc41e9f3a3b3634f21012bc37ef57a22f35638c93eca3f0ce1fd92223
    Input summary revisions
    Direct evidence analysis
  • GPT Astra ·
    Analysis stage
    evidence_segment
    Requested / returned model
    gpt-6-astra / Not supplied
    Exact provider snapshot
    Exact provider snapshot not supplied.
    Reasoning effort
    high
    Prompt version
    model-summary-v3
    Analysis input hash
    ab70d39786778cd386aaeeee6ae5c184de71f27bad1d2328c3999f3cd7824714
    Summary revision
    summary-9effc37473aa7d80951f9632e7098c1030a3a9cbdec18bef098bb769d702570b
    Input summary revisions
    Direct evidence analysis
  • GPT Astra ·
    Analysis stage
    example_analysis
    Requested / returned model
    gpt-6-astra / Not supplied
    Exact provider snapshot
    Exact provider snapshot not supplied.
    Reasoning effort
    high
    Prompt version
    model-summary-v3
    Analysis input hash
    e7aa1130f2fefa457fb1ef84de9c0d3f0695ccd39bb232549a9479fb718db366
    Summary revision
    summary-a54afafd0ab30c35e5506c89860b90fd7c9af896c15f4d239aba71e629db727d
    Input summary revisions
    Direct evidence analysis
  • GPT Astra ·
    Analysis stage
    suite_analysis
    Requested / returned model
    gpt-6-astra / Not supplied
    Exact provider snapshot
    Exact provider snapshot not supplied.
    Reasoning effort
    high
    Prompt version
    model-summary-v3
    Analysis input hash
    b8db26be71003bf7fce16ae743cfa0feda3aa0a2681e9f916f2f46843b420bce
    Summary revision
    summary-aee90cba6d16163810ae20b3705c1f61cdbb842407337a92c4e799f1a1541aa0
    Input summary revisions
    Direct evidence analysis
  • GPT Astra ·
    Analysis stage
    example_analysis
    Requested / returned model
    gpt-6-astra / Not supplied
    Exact provider snapshot
    Exact provider snapshot not supplied.
    Reasoning effort
    high
    Prompt version
    model-summary-v3
    Analysis input hash
    1e3a92add4d4d12a5da741a813583c0d1f4419e959241fe90541ae3895cbf61d
    Summary revision
    summary-b92d60790437cd19c0cabdf480089da6ee4f8b630a19517881081e45ad2abd7b
    Input summary revisions
    Direct evidence analysis
  • GPT Astra ·
    Analysis stage
    suite_analysis
    Requested / returned model
    gpt-6-astra / Not supplied
    Exact provider snapshot
    Exact provider snapshot not supplied.
    Reasoning effort
    high
    Prompt version
    model-summary-v3
    Analysis input hash
    4d27b6cad1175aa44b52735e34ab2f9a00180e5627bf993bd7235eb5ff6d8c56
    Summary revision
    summary-c6e38cfb1f4af7f348f665549dab55bd2afee1219e0a1ee9f686f20ec7403665
    Input summary revisions
    Direct evidence analysis
  • GPT Astra ·
    Analysis stage
    example_analysis
    Requested / returned model
    gpt-6-astra / Not supplied
    Exact provider snapshot
    Exact provider snapshot not supplied.
    Reasoning effort
    high
    Prompt version
    model-summary-v3
    Analysis input hash
    67db16fbe3b63b7c40a70c309bd65d2e9082223e18081417f8f91c02224c1ae6
    Summary revision
    summary-c7c37015114faf09b4730a5a8f1ff272e4bf432a8976b3a52ff3d4aa3d47f170
    Input summary revisions
    Direct evidence analysis
  • GPT Astra ·
    Analysis stage
    suite_synthesis
    Requested / returned model
    gpt-6-astra / Not supplied
    Exact provider snapshot
    Exact provider snapshot not supplied.
    Reasoning effort
    high
    Prompt version
    model-summary-v3
    Analysis input hash
    4dde65d5d71725664c43c08fba876c4a37abde90b038e559d6be4f683b605546
    Summary revision
    summary-deba09ea9bb8095260bd593b6b83a7d0fe0716e15da2a864cdad6c9f8ba1bc87
    Input summary revisions
    summary-072b62a90f03a8a897d9c1e5bec2c457dc1342b4b264fc2eb1abf1ccae8da10d, summary-0732052a3b0dd13cba04da2aad82f18250d5c966fd1c029887ee76029a4dde9c, summary-1962ce22c342d6b7181c7524b56f9cae2d508cb3cfddea5931ea4e1d3c67e800, summary-60fc6c2721b30bf1b93fce9cacfb79f0aae17b1b608af27364d92669273fb2d9, summary-76d5e82b6a8424978f0074829d5d9d2f879d3baabc5c993f30b811c88f38621c, summary-8030273b38146b827f397bd46e34cc296edb00ca2be4f3d68e781e01d5f5a101, summary-9b2c30abc41e9f3a3b3634f21012bc37ef57a22f35638c93eca3f0ce1fd92223, summary-9effc37473aa7d80951f9632e7098c1030a3a9cbdec18bef098bb769d702570b, summary-a54afafd0ab30c35e5506c89860b90fd7c9af896c15f4d239aba71e629db727d, summary-b92d60790437cd19c0cabdf480089da6ee4f8b630a19517881081e45ad2abd7b, summary-c7c37015114faf09b4730a5a8f1ff272e4bf432a8976b3a52ff3d4aa3d47f170, summary-ff14ac409b2944cd4086f5a43d3ba9037b4a7a8b0f2bf43fad26e12810ac4d12
  • GPT Astra ·
    Analysis stage
    evidence_segment
    Requested / returned model
    gpt-6-astra / Not supplied
    Exact provider snapshot
    Exact provider snapshot not supplied.
    Reasoning effort
    high
    Prompt version
    model-summary-v3
    Analysis input hash
    d7790ef493608d3b3182bf99b0ef9a780eb14ee6d7abfef6e0e81ea80e550568
    Summary revision
    summary-ff14ac409b2944cd4086f5a43d3ba9037b4a7a8b0f2bf43fad26e12810ac4d12
    Input summary revisions
    Direct evidence analysis

Good at

Handles small project edits and longer tool-driven work well with thinking off. It completed most larger practical tasks, making it worth considering for building small apps and investigating files—not just isolated coding exercises.

Evidence for this judgment
Everyday coding tasks · Thinking off
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 off
Average score90.9/100 · Mean task credit
Test coverage11 of 11 tasks scored
Task outcomes10 full credit · 0 partial · 1 no credit
Explore these test results →

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

Keeps source-backed answers when challenged and updates them when the source changes. With thinking off, it also solved most logic-and-number puzzles, showing useful reasoning beyond straightforward information lookup.

Evidence for this judgment
Resisting incorrect corrections · Thinking off
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 →

Logic and number puzzles · Thinking off
Average score82.4/100 · Mean puzzle credit
Test coverage34 of 34 tasks scored
Task outcomes28 full credit · 1 partial · 5 no credit
Explore these test results →

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

Struggles with

Research-paper answering was poor in both modes, not merely a citation-format problem: answer matching scores were low. Check answers against the paper yourself rather than treating this as a dependable research assistant.

Evidence for this judgment
Reading research papers · Thinking off
Average score24.8/100 · Answer token F1
Test coverage8 of 8 tasks scored
Task outcomes0 full credit · 4 partial · 4 no credit
Explore these test results →

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

Reading research papers · Thinking on
Average score10.4/100 · Answer token F1
Test coverage8 of 8 tasks scored
Task outcomes0 full credit · 3 partial · 5 no credit
Explore these test results →

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

Can get stuck revisiting a problem and omit the final answer. In the examples checked, thinking-enabled puzzle and formula-matching responses were cut off mid-analysis, leaving the requested result unfinished.

Evidence for this judgment

Example 3: Published hard problems
Thinking on. Final result: no credit. A final score does not describe every tool step; this example supports a behavior observation, not a whole-test conclusion.
View the question and recorded answer →

Example 4: Published hard problems
Thinking on. Final result: 10.0/100 · Mean answer credit. A final score does not describe every tool step; this example supports a behavior observation, not a whole-test conclusion.
View the question and recorded answer →

Should you enable thinking?

Start with thinking off for coding and practical work. Thinking slowed both, left baseline small-edit success unchanged, and worsened larger-task results. Separate sampled coding repetitions slightly favored thinking, so the default is a speed–quality choice, not an absolute rule.

Evidence for this judgment
Everyday coding tasks · Thinking off versus on
Thinking off → on100.0% · Task pass rate → 100.0% · Task pass rate
Paired test coverage16 of 16 tasks compared in both modes
Tasks timed in both modes16 pairs
Average time change with thinking on46.0 seconds slower
Explore these test results →

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

Larger practical tasks · Thinking off versus on
Thinking off → on90.9/100 · Mean task credit → 72.7/100 · Mean task credit
Paired test coverage11 of 11 tasks compared in both modes
Tasks timed in both modes11 pairs
Average time change with thinking on140.0 seconds slower
Explore these test results →

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

Everyday coding tasks · Thinking off · sampled repetitions
Repetitions earning full credit46 repetitions
Repetitions scored48 repetitions
Repetitions planned48 repetitions
Explore these test results →

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

Everyday coding tasks · Thinking on · sampled repetitions
Repetitions earning full credit48 repetitions
Repetitions scored48 repetitions
Repetitions planned48 repetitions
Explore these test results →

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

Consider thinking for passage questions or strict wording and formatting requirements: both improved, at the cost of longer waits. These gains concern short-passage answers and instruction compliance, not research-paper reading.

Evidence for this judgment
Reading without guessing · Thinking off versus on
Thinking off → on87.5% · Exact-match accuracy → 100.0% · Exact-match accuracy
Paired test coverage16 of 16 tasks compared in both modes
Score change with thinking on+12.5 score points
Average time change with thinking on19.8 seconds slower
Tasks timed in both modes16 pairs
Explore these test results →

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

Following exact instructions · Thinking off versus on
Thinking off → on82.6% · Strict prompt compliance → 91.3% · Strict prompt compliance
Paired test coverage23 of 23 tasks compared in both modes
Score change with thinking on+8.7 score points
Average time change with thinking on76.9 seconds slower
Tasks timed in both modes23 pairs
Explore these test results →

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

Worth knowing

Publication redactions, withheld artifacts and unavailable components limit reader inspection. Paper exports provide source links, not full text; exported project changes do not represent the original starting workspace.

Evidence for this judgment

Source limitation
Some selected evidence has recorded publication redactions.

Example 1: Larger practical tasks
Thinking on. Final result: no credit. A final score does not describe every tool step; this example supports a behavior observation, not a whole-test conclusion.
View the question and recorded answer →

Example 5: Larger practical tasks
Thinking off. Final result: full credit. A final score does not describe every tool step; this example supports a behavior observation, not a whole-test conclusion.
View the question and recorded answer →

Example 2: Everyday coding tasks
Thinking on. Final result: full credit. A final score does not describe every tool step; this example supports a behavior observation, not a whole-test conclusion.
View the question and recorded answer →

Model-authored project checks were not independently executed for this summary. The file-search failure lacks an explained criterion, and launcher checks do not establish that the desktop application ran.

Evidence for this judgment

Source limitation
Some selected evidence has recorded publication redactions.

Example 1: Larger practical tasks
Thinking on. Final result: no credit. A final score does not describe every tool step; this example supports a behavior observation, not a whole-test conclusion.
View the question and recorded answer →

Example 5: Larger practical tasks
Thinking off. Final result: full credit. A final score does not describe every tool step; this example supports a behavior observation, not a whole-test conclusion.
View the question and recorded answer →

Example 2: Everyday coding tasks
Thinking on. Final result: full credit. A final score does not describe every tool step; this example supports a behavior observation, not a whole-test conclusion.
View the question and recorded answer →

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

16/16 matched scored questions · 16/16 matched timed questions

Both modes earned full credit on every matched question. These matched results leave no measured quality headroom; unmatched questions are not covered by this observation.

Thinking offMean matched task credit100.0%
Thinking onMean matched task credit100.0%

Thinking on took 46.0 seconds more per matched timed question, on average.

All matched quality and timing measurements
Matched measurementRecorded result
Average credit, off → on100.0% → 100.0%
Mean paired credit change0.0 percentage points
Question outcomes0 improved · 16 unchanged · 0 worse
Mean added task time+46.0s
Mean task time, off → on17.9s → 63.9s
Median task time, off → on10.6s → 23.4s
Median of paired time changes+14.5s
Total added task time+736.1s
Ratio of summed times, on / off3.6×

Quality and timing can cover different subsets. Median paired change is calculated per question; it is not the difference between the two medians. Task time includes tools and grading.

Inspect the matched modes and their answers →
Inspect individual matched questions

Sampled repetitions

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

Thinking off · temperature 0.6

48/48 repeats scored · 46 full-credit repeats.

16/16 questions have every planned repeat scored. Of those, 15 earned full credit on every repeat.

Recorded repetition conditions
  • Seed 101 · requested off · verified
  • Seed 202 · requested off · verified
  • Seed 303 · requested off · verified

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.

Thinking on versus off

Verified requested on/off conditions of the same configuration, matched on the recorded test protocol.

Requested thinking on · Complete compared coverage

Compare these configurations →

Selection rule and evidence IDsPolicy comparisons-v3 · Template verified-thinking-pair
Facts: paired-a40ebdcb5dec0c2c7458cc78
Matching candidate set: paired-a40ebdcb5dec0c2c7458cc78

Qwen3.6 28B-REAP20-A3B

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

Requested thinking off · Complete compared coverage

Compare these configurations →

Selection rule and evidence IDsPolicy comparisons-v3 · Template same-parameter-architecture-matching-controls
Facts: fact-13b07e4fd98f9401f28d5b10, fact-16816c9c04532a86f188323f
Matching candidate set: fact-16816c9c04532a86f188323f, fact-9c25818e1688f1a4f03239a9, fact-b02c9416cc8ad504bf20376d, fact-47ad91563613ac587b7cf8d6, fact-5ae8ac173573895133028987, fact-fd3890f4fbd1383d2af80d72, fact-bfc465187e0c4b64d3e0faee, fact-84b68ee522c794052dbb13dd, fact-543885d1abe7569641164958, fact-d3eb63c2e0bc62fa106a3ff4, fact-3ca95ea92b5d2c9ac3e39bf6, fact-3bed0f728335765f1f64dc04, fact-f7565dbc26202571f4b7c51e, fact-2495f2a50aa1e8662cf535ad

Tested weights

UD-Q4_K_M · 16.95 GB

unsloth · recorded source

f2c28b3dc4776931ac6f879e11f203dec637ea0f14267a86ec8f6165f63f293f · Exact GGUF file →Historical first-MiB hash only; not a full-file identity.

Explore the answers

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