SapiensReleased Jul 24, 2026
Open weights

Agnes 2.5 Pro Alpha

Its best results are in Academic Reasoning (#80 of 223) and Reliability (#140 of 188). Five of seven dimensions are estimated, because no results are published yet.

Composite IQ ?
96
#155 of 186 ranked models · ahead of 17%
World (Other)'s #16 model, behind mistral-large-4 (118)
Effective cost
Not priced
No standard API rate
Speed
132tok/s
21s typical response
Context
1Mtokens
1,000,000 tokens in
Evidence
2of 7 dims
5 of 39 benchmarks

IQ profile

Seven dimensions, averaged equally into Composite IQ
Compare shape with
100 avg human130160● Abstract79#184 · est.● Math87#173 · est.● Academic125#80● Coding91#176 · est.● Computer Use93#147 · est.● Reliability105#140● Knowledge Work91#159 · est.
MeasuredEstimated95% range
#1 in dimensionTop-10 median

IQ dimensions

Ranks are among models scored in each dimension. Ranges are model-based estimates, not calibrated confidence intervals.

DimensionIQPosition · 70–160RankEvidencevs Top 10
79estimated
#184 / 221
0/5
-53.5
87estimated
#173 / 186
0/5
-62.5
125measured
#80 / 223
4/6
-10
91estimated
#176 / 187
0/6
-60
93estimated
#147 / 186
0/5
-53.5
105measured
#140 / 188
1/6
-30.5
91estimated
#159 / 186
0/6
-41
Agnes 2.5 Pro Alpha95% range (model-based, uncalibrated)Top-10 median#1 in dimension100 = average human

Benchmark results

5 of 39 Composite IQ benchmarks have a published result for Agnes 2.5 Pro Alpha. Benchmarks without a result stay listed so gaps are visible.

Abstract Reasoning IQ 79 · estimatedNo published results0/5
No published results yet for Abstract Reasoning. The 79 shown is estimated from Agnes 2.5 Pro Alpha's other dimensions and will be replaced as results come in. How estimates work
ARC-AGI-3ARC-AGI-2ARC-AGI-1Chess PuzzlesMystery Game Puzzles
Mathematical Reasoning IQ 87 · estimatedNo published results0/5
No published results yet for Mathematical Reasoning. The 87 shown is estimated from Agnes 2.5 Pro Alpha's other dimensions and will be replaced as results come in. How estimates work
FrontierMath Tier 4FrontierMath Tier 1-3ProofBench 1.1MathArenaAIME
Academic Reasoning IQ 125 · measuredBest: #58 of 159 on CritPt4/6
BenchmarkScorePercentileRankLeader · source
CritPt
Artificial Analysis model leaderboard · Sep 10
10.9%
#58 / 159
Leader Muse Wonder Full · 32.9%
Artificial Analysis model leaderboard · Sep 10
GPQA Diamond
Artificial Analysis model leaderboard · Sep 10
87.6%
#63 / 147
Leader gpt-6-astra · 96.1%
Artificial Analysis model leaderboard · Sep 10
Humanity's Last Exam
Artificial Analysis model leaderboard · Sep 10
33.6%
#76 / 167
Leader opus-5.5 · 61.4%
Artificial Analysis model leaderboard · Sep 10
SciCode
Artificial Analysis model leaderboard · Sep 10
42.2%
#108 / 167
Leader opus-5.5 · 66.9%
Artificial Analysis model leaderboard · Sep 10
Not yet reported
MMLU-ProMMMU-Pro
Programmatic Reasoning IQ 91 · estimatedNo published results0/6
No published results yet for Programmatic Reasoning. The 91 shown is estimated from Agnes 2.5 Pro Alpha's other dimensions and will be replaced as results come in. How estimates work
LiveCodeBenchIOI V2Terminal-Bench 4.0ProgramBenchSWE-rebenchMirrorCode
Computer Use IQ 93 · estimatedNo published results0/5
No published results yet for Computer Use. The 93 shown is estimated from Agnes 2.5 Pro Alpha's other dimensions and will be replaced as results come in. How estimates work
BrowseCompOSWorld-VerifiedMCP AtlasArena.ai Agent ArenaAgents' Last Exam
Reliability IQ 105 · measuredBest: #118 of 158 on AA Omniscience1/6
BenchmarkScorePercentileRankLeader · source
AA Omniscience
Artificial Analysis model leaderboard · Sep 10
-25.1
#118 / 158
Leader opus-5.5 · 46.4
Artificial Analysis model leaderboard · Sep 10
Not yet reported
SimpleQA VerifiedBullshitBench v2IFBenchAA Long Context Reasoning v1.1FACTS Grounding
Knowledge Work IQ 91 · estimatedNo published results0/6
No published results yet for Knowledge Work. The 91 shown is estimated from Agnes 2.5 Pro Alpha's other dimensions and will be replaced as results come in. How estimates work
AA-Briefcase v1.1GDPval-AA v2.1AutomationBench-AAGDP.pdfAA-AnalystAgentEnterpriseOps-Gym-AA

Release & availability

Facts from the publisher's announcement and documentation.

Released
Developer
Sapiens
Access
Official downloadable open weights under Apache 2.0; the Agnes AI hosted Alpha API is deprecated.
Weights
Open weightsDownloadable and self-hostable
Inputs
textimage
Context
1,000,000 tokens
API price
No per-token rateNo current standard hosted token price has been verified for this checkpoint; downloadable weights are not a zero-dollar hosted API.

What Sapiens says

  • Agnes AI released this text-and-image reasoning checkpoint on July 24, 2026.
  • Official weights remain available under Apache 2.0; the historical hosted endpoint is deprecated.

Limitations

  • Hosted agnes-2.5-pro-alpha API is deprecated; use the current model for new hosted integrations. Open weights remain available.
  • Historical Alpha hosted prices are not current billable prices.

Publisher statements, not AI IQ measurements.

About this score

How much of Agnes 2.5 Pro Alpha's IQ is measured, and how the rest is estimated.

5/ 39benchmark results
2/ 7dimensions measured
5estimated
0/5Abstract
0/5Math
4/6Academic
0/6Coding
0/5Computer Use
1/6Reliability
0/6Knowledge Work
Bar height = share of benchmarks with resultsEstimatedNo score
Measured results are shown as published

Every score in the benchmark tables comes from its source unchanged. Composite IQ averages all seven dimensions equally.

Estimates fill the gaps

Abstract Reasoning, Mathematical Reasoning, Programmatic Reasoning, Computer Use and Knowledge Work have no published results yet, so their IQ is estimated from Agnes 2.5 Pro Alpha's other dimensions.

A missing result is not a zero

Benchmark versions and evaluation harnesses can differ between sources, so small gaps between models may not be meaningful.

Read the scoring methodology →

Compare Agnes 2.5 Pro Alpha

Side-by-side pages for the models it is most often weighed against.

Agnes 2.5 Pro AlphaComparison model
Browse all comparisons →

Frequently asked questions

Quick answers about Agnes 2.5 Pro Alpha, drawn from the same data as the rest of this page.

How AI IQ scores models →Get this data from the API →
What is the IQ of Agnes 2.5 Pro Alpha?
Agnes 2.5 Pro Alpha has an estimated IQ of 96, which ranks #155 out of 186 public models on the AI IQ leaderboard. The score combines seven equally weighted reasoning dimensions using measured benchmark results and conservative estimates for missing coverage.
What is Agnes 2.5 Pro Alpha best at?
Among the dimensions with published benchmark results, Agnes 2.5 Pro Alpha ranks highest in Academic Reasoning (IQ 125, #80 of 223) and Reliability (IQ 105, #140 of 188).
What is the EQ of Agnes 2.5 Pro Alpha?
Agnes 2.5 Pro Alpha doesn't have an EQ score yet. EQ (emotional reasoning) is calculated from EQ-Bench 4 and AttuneBench, and neither has published a result for Agnes 2.5 Pro Alpha. AI IQ doesn't estimate EQ from other benchmarks, so the score appears once one of those results is published.
Which benchmarks has Agnes 2.5 Pro Alpha been tested on?
Agnes 2.5 Pro Alpha has published results on 5 of the 39 benchmarks behind AI IQ's Composite IQ, including CritPt, GPQA Diamond, Humanity's Last Exam, SciCode and AA Omniscience.
How much does Agnes 2.5 Pro Alpha cost?
AI IQ doesn't have a verified standard per-token API price for Agnes 2.5 Pro Alpha, so it can't calculate an effective cost. This is common for models offered only through subscriptions, restricted-access programs, or self-hosting.
Is Agnes 2.5 Pro Alpha open source?
Yes — Agnes 2.5 Pro Alpha is an open-weights model: its trained weights are published, so it can be downloaded and self-hosted.
Who developed Agnes 2.5 Pro Alpha?
Agnes 2.5 Pro Alpha was developed by Sapiens, based in Singapore.
When was Agnes 2.5 Pro Alpha released?
Agnes 2.5 Pro Alpha was publicly released on July 24, 2026.
How large is the context window of Agnes 2.5 Pro Alpha?
Agnes 2.5 Pro Alpha supports a context window of 1,000,000 tokens — the amount of text it can consider in a single request.
How fast is Agnes 2.5 Pro Alpha?
In independent measurements, Agnes 2.5 Pro Alpha generates about 132 tokens per second, with a typical response time of about 21 seconds.