GoogleReleased Apr 2, 2026
Open weights
gemma-4-26b
Its best results are in Reliability (#103 of 177) and Academic Reasoning (#130 of 203); weakest in Knowledge Work (#114 of 176). Four of seven dimensions are estimated, because no results are published yet.
Composite IQ ?
99
Effective cost
$0.137/1M
$0.10 in · $0.37 out
Speed
94tok/s
27s typical response
Context
256Ktokens
256,000 tokens in
Evidence
3of 7 dims
13 of 39 benchmarks
IQ profile
Seven dimensions, averaged equally into Composite IQ
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
gemma-4-26b95% range (model-based, uncalibrated)Top-10 median#1 in dimension100 = average human
Benchmark results
13 of 39 Composite IQ benchmarks have a published result for gemma-4-26b. Benchmarks without a result stay listed so gaps are visible.
Abstract Reasoning IQ 86 · estimatedNo published results0/5
No published results yet for Abstract Reasoning. The 86 shown is estimated from gemma-4-26b'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 92 · estimatedNo published results0/5
No published results yet for Mathematical Reasoning. The 92 shown is estimated from gemma-4-26b'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 112 · measuredBest: #103 of 147 on GPQA Diamond5/6
BenchmarkScorePercentileRankLeader · source
GPQA Diamond
Artificial Analysis v4.3.2 · Oct 27
79.2%
#103 / 147
Leader gpt-6-astra · 96.1%
Artificial Analysis v4.3.2 · Oct 27
Artificial Analysis v4.3.2 · Oct 27
Humanity's Last Exam
Artificial Analysis v4.3.2 · Oct 27
19.3%
#121 / 167
Leader opus-5.5 · 61.4%
Artificial Analysis v4.3.2 · Oct 27
Artificial Analysis v4.3.2 · Oct 27
SciCode
Artificial Analysis v4.3.2 · Oct 27
40.0%
#122 / 167
Leader opus-5.5 · 66.9%
Artificial Analysis v4.3.2 · Oct 27
Artificial Analysis v4.3.2 · Oct 27
MMMU-Pro
Artificial Analysis v4.3.2 · Oct 27
69.2%
#69 / 89
Leader fable-5.1 · 90.6%
Artificial Analysis v4.3.2 · Oct 27
Artificial Analysis v4.3.2 · Oct 27
CritPt
Artificial Analysis v4.3.2 · Oct 27
0.0%
#139 / 159
Leader Muse Wonder Full · 32.9%
Artificial Analysis v4.3.2 · Oct 27
Artificial Analysis v4.3.2 · Oct 27
Not yet reported
MMLU-Pro
Programmatic Reasoning IQ 97 · estimatedNo published results0/6
No published results yet for Programmatic Reasoning. The 97 shown is estimated from gemma-4-26b's other dimensions and will be replaced as results come in. How estimates work
LiveCodeBenchIOI V2Terminal-Bench 4.0ProgramBenchSWE-rebenchMirrorCode
Computer Use IQ 98 · estimatedNo published results0/5
No published results yet for Computer Use. The 98 shown is estimated from gemma-4-26b's other dimensions and will be replaced as results come in. How estimates work
BrowseCompOSWorld-VerifiedMCP AtlasArena.ai Agent ArenaAgents' Last Exam
Reliability IQ 110 · measuredBest: #4 of 36 on FACTS Grounding5/6
BenchmarkScorePercentileRankLeader · source
FACTS Grounding
Full source leaderboard reconciliation · Oct 23
80.9%
#4 / 36
Leader gpt-6-sol · 85.2%
Full source leaderboard reconciliation · Oct 23
Full source leaderboard reconciliation · Oct 23
IFBench
Artificial Analysis v4.3.2 · Oct 27
72.4%
#36 / 96
Leader minimax-m3 · 82.9%
Artificial Analysis v4.3.2 · Oct 27
Artificial Analysis v4.3.2 · Oct 27
BullshitBench v2
Full source leaderboard reconciliation · Oct 23
21.0%
#63 / 83
Leader opus-4.8 · 94.0%
Full source leaderboard reconciliation · Oct 23
Full source leaderboard reconciliation · Oct 23
AA Long Context Reasoning v1.1
Artificial Analysis v4.3.2 · Oct 27
65.7%
#120 / 154
Leader kimi-k3 · 88.7%
Artificial Analysis v4.3.2 · Oct 27
Artificial Analysis v4.3.2 · Oct 27
AA Omniscience
Artificial Analysis v4.3.2 · Oct 27
-50.8
#149 / 158
Leader opus-5.5 · 46.4
Artificial Analysis v4.3.2 · Oct 27
Artificial Analysis v4.3.2 · Oct 27
Not yet reported
SimpleQA Verified
Knowledge Work IQ 100 · measuredBest: #101 of 124 on GDPval-AA v2.1 normalized score3/6
BenchmarkScorePercentileRankLeader · source
GDPval-AA v2.1 normalized score
Artificial Analysis v4.3.2 · Oct 23
3.3%
#101 / 124
Leader opus-5.5 · 68.3%
Artificial Analysis v4.3.2 · Oct 23
Artificial Analysis v4.3.2 · Oct 23
AutomationBench-AA
Artificial Analysis v4.3.2 · Oct 27
1.9%
#89 / 106
Leader Gemini 4 Argon · 77.5%
Artificial Analysis v4.3.2 · Oct 27
Artificial Analysis v4.3.2 · Oct 27
AA-Briefcase v1.1 Elo
Artificial Analysis v4.3.2 · Oct 27
315
#97 / 111
Leader sonnet-5.5 · 1,812
Artificial Analysis v4.3.2 · Oct 27
Artificial Analysis v4.3.2 · Oct 27
Not yet reported
GDP.pdfAA-AnalystAgentEnterpriseOps-Gym-AA
Release & availability
Release details on record for this model.
- Released
- Developer
- Weights
- Open weightsDownloadable and self-hostable
- Context
- 256,000 tokens
- API price
- $0.10 in · $0.37 outper 1M tokens · full pricing
About this score
How much of gemma-4-26b's IQ is measured, and how the rest is estimated.
13/ 39benchmark results
3/ 7dimensions measured
4estimated
0/5Abstract
0/5Math
5/6Academic
0/6Coding
0/5Computer Use
5/6Reliability
3/6Knowledge 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 and Computer Use have no published results yet, so their IQ is estimated from gemma-4-26b's other dimensions. Missing benchmarks can also draw on its predecessor, Gemma 4 31b; a predecessor is an estimation input, not evidence that this model has the predecessor's measured results.
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.
Compare gemma-4-26b
Side-by-side pages for the models it is most often weighed against.
#1 overall
gpt-6.1-sol
IQ141 42 higherEffective cost$6.30 46.0× pricier
Predecessor
gemma-4-31b
IQ101 2 higherEffective cost$0.137 about the same
Google #1
Gemini 4 Argon
IQ140 41 higher
Closest IQ
hy3
IQ99 same IQEffective cost$0.654 4.8× pricier
Most similar profile
Mercury 3
IQ98 1 lowerEffective cost$0.403 2.9× pricier
Similar price
Qwen3.7 Flash
IQ94 5 lowerEffective cost$0.149 about the same
gemma-4-26bComparison model
Frequently asked questions
Quick answers about gemma-4-26b, 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 gemma-4-26b?
- gemma-4-26b has an estimated IQ of 99, which ranks #131 out of 176 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 gemma-4-26b best at?
- Among the dimensions with published benchmark results, gemma-4-26b ranks highest in Reliability (IQ 110, #103 of 177) and Academic Reasoning (IQ 112, #130 of 203). Its lowest-ranked measured dimension is Knowledge Work (IQ 100, #114 of 176).
- What is the EQ of gemma-4-26b?
- gemma-4-26b has an estimated EQ of 108. EQ measures emotional reasoning — how well a model reads and responds to emotional context — and is scored separately from IQ.
- Which benchmarks has gemma-4-26b been tested on?
- gemma-4-26b has published results on 13 of the 39 benchmarks behind AI IQ's Composite IQ, including FACTS Grounding, IFBench, GPQA Diamond, Humanity's Last Exam and SciCode.
- How much does gemma-4-26b cost?
- gemma-4-26b costs $0.10 per million input tokens and $0.37 per million output tokens. Adjusted for how many tokens it actually consumes on real tasks, its effective cost works out to $0.1371 per million input-plus-output tokens.
- Is gemma-4-26b open source?
- Yes — gemma-4-26b is an open-weights model: its trained weights are published, so it can be downloaded and self-hosted.
- Who developed gemma-4-26b?
- gemma-4-26b was developed by Google, based in the United States.
- When was gemma-4-26b released?
- gemma-4-26b was publicly released on April 2, 2026.
- How large is the context window of gemma-4-26b?
- gemma-4-26b supports a context window of 256,000 tokens — the amount of text it can consider in a single request.
- How fast is gemma-4-26b?
- In independent measurements, gemma-4-26b generates about 94 tokens per second, with a typical response time of about 27 seconds.