UAB
Home/Find
Find
4 stale sources

Describe the job.

Set filters, get ranked options with verified scores and source dates attached.
Recommendation · Open-weight shortlist

Best evidence-backed choice for Open-weight shortlist with open models only: Qwen3.8 2.4T A95B.

Qwen3.8 2.4T A95BQwen · mid
Verified & stable

The main warning is thinner evidence on Coding.

Direct sources
16
Coverage
75%
Data version
Oct 3, 2026
Compare top optionsShare this report
Evidence behind the call
Data version Oct 3, 202616 direct source linksNo excluded sourcesAll public sources
What the data can support5 ready, 1 partial, 1 missing for the current task result.
Minimum source requirementenough direct data · enough direct data · enough direct data · enough direct data
Cost / latencyLatency is unavailable in the current verified source data.
Best pickQwen3.8 2.4T A95B
Runner-upQwen3.8 Max
Coverage75% visible · 75% checked
PresetOpen-weight shortlist
SourcesAll public sources
Latest strong source dataOct 3, 2026
Where sources differ4 source rows behind this answer
1
Qwen3.8 2.4T A95BArtificial Analysis · GPQA · 93.535 · #15 · 98pctl
used for answer
2
Qwen3.8 2.4T A95BArtificial Analysis · Intelligence Index · 39.886 · #50 · 91pctl
used for answer
3
Qwen3.8 2.4T A95BArtificial Analysis · CritPt · 20 · #44 · 91pctl
used for answer
4
Qwen3.8 2.4T A95BArtificial Analysis · Humanity's Last Exam · 42.447 · #51 · 91pctl
used for answer
Compare top optionsOpen comparison tableCoverage report
Ranked options

Direct matches first, then weaker or missing-data cases

Direct matches stay strict; strong models with indirect data still surface below. Open a row for its scores, source links, and caveats.

Primary group · Direct-match leaders
#1Qwen3.8 2.4T A95BQwen · mid89.375% visible
enough direct data · recent · strong · clearmid-priced price band from registry metadataLatency is unavailable in the current verified source data.
Verified and stable19.3% score spread · 100% recent data · exact alias
Fit score89.3
Strongest source datareasoning math science · chat text

Qwen3.8 2.4T A95B is strongest on Reasoning / math / science and Chat / text for this preset.

  • Last updated: Latest visible source row is 1 day old.
  • Where sources differ: 19-point cross-domain spread; warning threshold is 30.
  • Data checks: No open review item was matched to this recommendation.

The main warning is thinner evidence on Coding.

Verified rows
7
Hand-checked rows
0
Copied rows
0
Backfilled rows
0
Headline lane
Oct 3, 2026
Background data
No extra context
Formula recommendation-fit-v2.0.0.
Source linksArtificial AnalysisGPQA · 93.535 · #15 · 98pctlArtificial AnalysisIntelligence Index · 39.886 · #50 · 91pctlArtificial AnalysisCritPt · 20 · #44 · 91pctl
Measured by
public third-party sources
Source basis
exact alias
Compared fairly because
same test setup, version groups, source-balanced before averaging
Open model
Open compare
#2Qwen3.8 MaxQwen · mid87.775% visible
enough direct data · recent · strong · clearmid-priced price band from registry metadataLatency is unavailable in the current verified source data.
Verified and stable22% score spread · 100% recent data · exact alias
Fit score87.7
Strongest source datareasoning math science · chat text

Qwen3.8 Max is strongest on Reasoning / math / science and Chat / text for this preset.

  • Last updated: Latest visible source row is 1 day old.
  • Where sources differ: 22-point cross-domain spread; warning threshold is 30.
  • Data checks: No open review item was matched to this recommendation.

The main warning is thinner evidence on Coding.

Verified rows
7
Hand-checked rows
0
Copied rows
0
Backfilled rows
0
Headline lane
Oct 3, 2026
Background data
No extra context
Formula recommendation-fit-v2.0.0.
Source linksArtificial AnalysisGPQA · 92.727 · #27 · 95pctlArtificial AnalysisIntelligence Index · 40.153 · #49 · 91pctlArtificial AnalysisHumanity's Last Exam · 43.049 · #48 · 91pctl
Measured by
public third-party sources
Source basis
exact alias
Compared fairly because
same test setup, version groups, source-balanced before averaging
Open model
Open compare
#3qwen3.8-maxQwen · mid85.075% visible
enough direct data · recent · strong · clearmid-priced price band from registry metadataLatency is unavailable in the current verified source data.
Verified and stable14.4% score spread · 73.3% recent data · exact alias
Fit score85.0
Strongest source datachat text · reasoning math science

qwen3.8-max is strongest on Chat / text and Reasoning / math / science for this preset.

  • Last updated: Latest visible source row is 2 days old.
  • Where sources differ: 14-point cross-domain spread; warning threshold is 30.
  • Data checks: No open review item was matched to this recommendation.

The main warning is thinner evidence on Coding.

Verified rows
69
Hand-checked rows
0
Copied rows
0
Backfilled rows
0
Headline lane
Oct 3, 2026
Background data
No extra context
Formula recommendation-fit-v2.0.0.
Source linksLiveBenchZebra puzzle · 100 · #9 · 100pctlLiveBenchSpatial · 100 · #12 · 100pctlArenaText Arena · Hard Prompts English · No Style Control · 1,504.775 · #12 · 97pctl
Measured by
public third-party sources
Source basis
exact alias
Compared fairly because
same test setup, version groups, source-balanced before averaging
Open model
Open compare
#4qwen3.7-plusQwen · mid81.675% visible
enough direct data · recent · strong · clearmid-priced price band from registry metadataLatency is unavailable in the current verified source data.
Verified and stable24.1% score spread · 100% recent data · exact alias
Fit score81.6
Strongest source datareasoning math science · chat text

qwen3.7-plus is strongest on Reasoning / math / science and Chat / text for this preset.

  • Last updated: Latest visible source row is 1 day old.
  • Where sources differ: 24-point cross-domain spread; warning threshold is 30.
  • Data checks: No open review item was matched to this recommendation.

The main warning is thinner evidence on Coding.

Verified rows
34
Hand-checked rows
0
Copied rows
0
Backfilled rows
0
Headline lane
Oct 3, 2026
Background data
No extra context
Formula recommendation-fit-v2.0.0.
Source linksArtificial AnalysisIFBench · 77.959 · #7 · 98pctlArtificial AnalysisTerminal-Bench Hard · 46.97 · #10 · 97pctlArtificial AnalysisAA-Omniscience non-hallucination · 72.338 · #25 · 95pctl
Measured by
public third-party sources
Source basis
exact alias
Compared fairly because
same test setup, version groups, source-balanced before averaging
Open model
Open compare
#5qwen3.8-flash-nextQwen · mid74.775% visible
Source status unavailablemid price bandLatency unavailable
Verified but aging6.8% score spread · 51.5% recent data · exact alias
Fit score74.7
Strongest source datachat text · reasoning math science

qwen3.8-flash-next is strongest on Chat / text and Reasoning / math / science for this preset.

The verified evidence is decent, but too much of it is aging to treat this as a clean current winner.

Verified rows
36
Hand-checked rows
0
Copied rows
0
Backfilled rows
0
Headline lane
Oct 3, 2026
Background data
No extra context
Formula recommendation-fit-v2.0.0.

No source link clears the minimum source requirement.

Measured by
public third-party sources
Source basis
exact alias
Compared fairly because
same test setup, version groups, source-balanced before averaging
Open model
Open compare
Evidence & limits

Read the argument before you commit

Why this is the top optiontop reasons behind the current answer
  • Qwen3.8 2.4T A95B leads this preset with a 89.3 weighted fit score.
  • Its strongest visible evidence comes from Reasoning / math / science and Chat / text.
  • Coverage sits at 75%, with 75% backed by verified runtime or manual verification.
What to pressure testwhere the current answer is still fragile
  • Runner-up edge · Qwen3.8 Max: Qwen3.8 Max is strongest on Reasoning / math / science and Chat / text for this preset.
  • Evidence risk: The main warning is thinner evidence on Coding.
What would flip the answerthe assumptions the result rests on
  • If you tighten benchmark spread: Qwen3.8 2.4T A95B still holds if you care more about aligned evidence than upside.
  • If you tighten recency: Qwen3.8 2.4T A95B remains viable because the visible evidence is still fairly fresh.
  • If cost and speed matter more: Qwen3.8 2.4T A95B
Why this is not a clean winlimitations to keep in mind
  • The main warning is thinner evidence on Coding.
  • Qwen3.8 Max remains close enough that a different scoring recipe can still flip the public answer.
Source links6 rows behind this answer
Data limitswhere the public data is thin
  • cost: partial · Only registry price bands are ready; exact price calculations are not shown.
  • latency: missing · Latency is not available in the current verified shortlist data.
Not chosen6 well-known models left off

qwen3.8-27b

Closest option

  • Missing benchmark coverage in Embeddings / retrieval.
  • qwen3.8-27b has direct evidence on part of this preset, but not enough to clear the exact-match floor.
Open model

qwen3.6-plus

Closest option

  • Missing benchmark coverage in Embeddings / retrieval.
  • qwen3.6-plus has direct evidence on part of this preset, but not enough to clear the exact-match floor.
Open model

Qwen3.6 27B (Reasoning)

Closest option

  • Missing benchmark coverage in Embeddings / retrieval.
  • Qwen3.6 27B (Reasoning) has direct evidence on part of this preset, but not enough to clear the exact-match floor.
Open model

Gemma 4 26B A4B

Closest option

  • Missing benchmark coverage in Embeddings / retrieval.
  • Gemma 4 26B A4B has direct evidence on part of this preset, but not enough to clear the exact-match floor.
Open model

gemma-4-31b

Closest option

  • Missing benchmark coverage in Embeddings / retrieval.
  • gemma-4-31b has direct evidence on part of this preset, but not enough to clear the exact-match floor.
Open model

Claude Fable 5

Known current model

  • Current generated catalog does not have enough matching source links for this task preset.
Open model
Needs more source data12 tracked models with thin public data

qwen3.8-27b

Qwen · 75% visible · 75% direct · 0% indirect

qwen3.8-27b has direct evidence on part of this preset, but not enough to clear the exact-match floor.

  • Missing benchmark coverage in Embeddings / retrieval.
codingchat text
Open model

qwen3.6-plus

Qwen · 75% visible · 75% direct · 0% indirect

qwen3.6-plus has direct evidence on part of this preset, but not enough to clear the exact-match floor.

  • Missing benchmark coverage in Embeddings / retrieval.
chat textcoding
Open model

Qwen3.6 27B (Reasoning)

Qwen · 75% visible · 75% direct · 0% indirect

Qwen3.6 27B (Reasoning) has direct evidence on part of this preset, but not enough to clear the exact-match floor.

  • Missing benchmark coverage in Embeddings / retrieval.
reasoning math sciencechat text
Open model

Gemma 4 26B A4B

Google · 75% visible · 75% direct · 0% indirect

Gemma 4 26B A4B has direct evidence on part of this preset, but not enough to clear the exact-match floor.

  • Missing benchmark coverage in Embeddings / retrieval.
reasoning math sciencechat text
Open model

gemma-4-31b

Google · 75% visible · 75% direct · 0% indirect

gemma-4-31b has direct evidence on part of this preset, but not enough to clear the exact-match floor.

  • Missing benchmark coverage in Embeddings / retrieval.
reasoning math sciencechat text
Open model

Mistral Large 3

Mistral · 75% visible · 75% direct · 0% indirect

Mistral Large 3 has direct evidence on part of this preset, but not enough to clear the exact-match floor.

  • Missing benchmark coverage in Embeddings / retrieval.
chat textcoding
Open model

Qwen3 Max

Qwen · 75% visible · 75% direct · 0% indirect

Qwen3 Max has direct evidence on part of this preset, but not enough to clear the exact-match floor.

  • Missing benchmark coverage in Embeddings / retrieval.
chat textreasoning math science
Open model

Qwen3.5 122B A10B

Qwen · 75% visible · 75% direct · 0% indirect

Qwen3.5 122B A10B has direct evidence on part of this preset, but not enough to clear the exact-match floor.

  • Missing benchmark coverage in Embeddings / retrieval.
codingreasoning math science
Open model

Qwen3.5 397B A17B

Qwen · 75% visible · 75% direct · 0% indirect

Qwen3.5 397B A17B has direct evidence on part of this preset, but not enough to clear the exact-match floor.

  • Missing benchmark coverage in Embeddings / retrieval.
codingreasoning math science
Open model

DeepSeek Chat

DeepSeek · 75% visible · 75% direct · 0% indirect

DeepSeek Chat has direct evidence on part of this preset, but not enough to clear the exact-match floor.

  • Missing benchmark coverage in Embeddings / retrieval.
codingreasoning math science
Open model

DeepSeek Reasoner

DeepSeek · 75% visible · 75% direct · 0% indirect

DeepSeek Reasoner has direct evidence on part of this preset, but not enough to clear the exact-match floor.

  • Missing benchmark coverage in Embeddings / retrieval.
chat textreasoning math science
Open model

Qwen3-Coder 480B A35B

Qwen · 75% visible · 75% direct · 0% indirect

Qwen3-Coder 480B A35B has direct evidence on part of this preset, but not enough to clear the exact-match floor.

  • Missing benchmark coverage in Embeddings / retrieval.
chat textcoding
Open model

Shareable claims with evidence

The product should generate public claims worth checking, not just filter state.

Open change report
alert
13 review items still need manual judgment

The product keeps parser and mapping ambiguity visible instead of silently guessing.

Open
sources
Artificial Analysis moved via source updated leaderboard

The saved raw source snapshot changed relative to the previous run. Window: 2026-10-03T11:41:25Z -> 2026-10-03T11:42:08Z.

Open
models
BridgeBench moved via new benchmark coverage

10 benchmark rows were added, 10 removed, and 0 existing rows changed value or evaluation date. Window: 2026-05-15T16:26:59Z -> 2026-05-15T16:34:46Z.

Open
sources
Scale Labs moved via source updated leaderboard

The saved raw source snapshot changed relative to the previous run. Window: 2026-10-03T11:41:42Z -> 2026-10-03T11:42:34Z.

Open
product
Initial comparison-table release

Added comparison-table homepage, same-test normalization, per-cell source links, source pages, and custom-ranking preview.

Open

What changed this week

alert
13 review items still need manual judgment

The product keeps parser and mapping ambiguity visible instead of silently guessing.

Open
sources
Artificial Analysis moved via source updated leaderboard

The saved raw source snapshot changed relative to the previous run. Window: 2026-10-03T11:41:25Z -> 2026-10-03T11:42:08Z.

Source-data window: 2026-10-03T11:41:25Z -> 2026-10-03T11:42:08Z

Open
models
BridgeBench moved via new benchmark coverage

10 benchmark rows were added, 10 removed, and 0 existing rows changed value or evaluation date. Window: 2026-05-15T16:26:59Z -> 2026-05-15T16:34:46Z.

Source-data window: 2026-05-15T16:26:59Z -> 2026-05-15T16:34:46Z

Open reportModelBenchmarkSource
sources
Scale Labs moved via source updated leaderboard

The saved raw source snapshot changed relative to the previous run. Window: 2026-10-03T11:41:42Z -> 2026-10-03T11:42:34Z.

Source-data window: 2026-10-03T11:41:42Z -> 2026-10-03T11:42:34Z

Open
product
Initial comparison-table release

Added comparison-table homepage, same-test normalization, per-cell source links, source pages, and custom-ranking preview.

Source-data window: 2026-04-16

Open changelog
models
Methodology contract published

Documented comparability rules, raw-vs-normalized behavior, and why unlike metrics are never averaged by default.

Source-data window: 2026-04-16

models
Artificial Analysis ID rule adopted

Stable model and creator IDs are now the preferred external identity keys when available.

Source-data window: 2026-04-15

models
BridgeBench parser fallback added

Added alternate selectors for category headers after leaderboard markup drift.

Source-data window: 2026-04-15

Data version

Current snapshot.

Published Oct 3, 2026Model list checked4 stale sources105 claim warnings9 providers · 1334 tracked models

Freshness is calculated at request time. Failed sources retain their last verified rows with a visible age penalty. View live data status.

Quick routes

Jump straight to a page.

Resolve a recommendation into a public reportbest open model for long-context researchResearch assistantOpen page
Send a shortlist into compare modecompare gpt-5, claude opus, gemini proEveryday chatbotOpen page
Open a head-to-head debate pagegpt-5 vs claude opusEveryday chatbotOpen page
Open a source-difference reportbenchmark controversy for livebench codingCoding copilotOpen page
Open the latest public movementwhat changed this weekEveryday chatbotOpen page
Jump straight to an entity pageopen model gpt-5Open-weight shortlistOpen page