Ant GroupReleased Aug 27, 2026
Open weights

Ling 3.0 Flash Fin

Its best results are in Academic Reasoning (#117 of 223) and Reliability (#99 of 188); weakest in Programmatic Reasoning (#140 of 187). Three of seven dimensions are estimated, because no results are published yet.

Composite IQ ?
102
#124 of 186 ranked models · ahead of 34%
China (Other)'s #10 model, behind Ling 3.1 Flash (120)
Effective cost
Not priced
No standard API rate
Speed
338tok/s
9s typical response
Context
262Ktokens
262,144 tokens in
Evidence
4of 7 dims
8 of 39 benchmarks

IQ profile

Seven dimensions, averaged equally into Composite IQ
Compare shape with
100 avg human130160● Abstract89#129 · est.● Math92#163 · est.● Academic118#117● Coding99#140● Computer Use97#135 · est.● Reliability111#99● Knowledge Work105#112
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
89estimated
#129 / 221
0/5
-43.5
92estimated
#163 / 186
0/5
-57.5
118measured
#117 / 223
3/6
-17
99measured
#140 / 187
1/6
-52
97estimated
#135 / 186
0/5
-49.5
111measured
#99 / 188
2/6
-24.5
105measured
#112 / 186
2/6
-27
Ling 3.0 Flash Fin95% range (model-based, uncalibrated)Top-10 median#1 in dimension100 = average human

Benchmark results

8 of 39 Composite IQ benchmarks have a published result for Ling 3.0 Flash Fin. Benchmarks without a result stay listed so gaps are visible.

Abstract Reasoning IQ 89 · estimatedNo published results0/5
No published results yet for Abstract Reasoning. The 89 shown is estimated from Ling 3.0 Flash Fin'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 Ling 3.0 Flash Fin'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 118 · measuredBest: #99 of 159 on CritPt3/6
BenchmarkScorePercentileRankLeader · source
CritPt
Artificial Analysis v4.3.2 · Oct 5
2.6%
#99 / 159
Leader Muse Wonder Full · 32.9%
Artificial Analysis v4.3.2 · Oct 5
SciCode
Artificial Analysis v4.3.2 · Oct 5
42.4%
#105 / 167
Leader opus-5.5 · 66.9%
Artificial Analysis v4.3.2 · Oct 5
Humanity's Last Exam
Artificial Analysis v4.3.2 · Oct 5
22.6%
#110 / 167
Leader opus-5.5 · 61.4%
Artificial Analysis v4.3.2 · Oct 5
Not yet reported
GPQA DiamondMMLU-ProMMMU-Pro
Programmatic Reasoning IQ 99 · measuredBest: #90 of 109 on Terminal-Bench 4.01/6
BenchmarkScorePercentileRankLeader · source
Terminal-Bench 4.0
Artificial Analysis v4.3.2 · Oct 5
0.0%
#90 / 109
Leader Kimi K3.1 Pro · 67.2%
Artificial Analysis v4.3.2 · Oct 5
Not yet reported
LiveCodeBenchIOI V2ProgramBenchSWE-rebenchMirrorCode
Computer Use IQ 97 · estimatedNo published results0/5
No published results yet for Computer Use. The 97 shown is estimated from Ling 3.0 Flash Fin's other dimensions and will be replaced as results come in. How estimates work
BrowseCompOSWorld-VerifiedMCP AtlasArena.ai Agent ArenaAgents' Last Exam
Reliability IQ 111 · measuredBest: #94 of 154 on AA Long Context Reasoning v1.12/6
BenchmarkScorePercentileRankLeader · source
AA Long Context Reasoning v1.1
Artificial Analysis v4.3.2 · Oct 5
73.7%
#94 / 154
Leader kimi-k3 · 88.7%
Artificial Analysis v4.3.2 · Oct 5
AA Omniscience
Artificial Analysis v4.3.2 · Oct 5
-14.6
#101 / 158
Leader opus-5.5 · 46.4
Artificial Analysis v4.3.2 · Oct 5
Not yet reported
SimpleQA VerifiedBullshitBench v2IFBenchFACTS Grounding
Knowledge Work IQ 105 · measuredBest: #73 of 112 on AutomationBench-AA2/6
BenchmarkScorePercentileRankLeader · source
AutomationBench-AA
Artificial Analysis v4.3.2 · Oct 5
7.4%
#73 / 112
Leader Gemini 4 Carbon · 77.8%
Artificial Analysis v4.3.2 · Oct 5
GDP.pdf
Artificial Analysis v4.3.2 · Oct 5
5.4%
#84 / 111
Leader Muse Wonder Full · 32.6%
Artificial Analysis v4.3.2 · Oct 5
Not yet reported
AA-Briefcase v1.1GDPval-AA v2.1AA-AnalystAgentEnterpriseOps-Gym-AA

Release & availability

Facts from the publisher's announcement and documentation.

Released
Developer
Ant Group
Access
MIT open weights and hosted API access.
Weights
Open weightsDownloadable and self-hostable
Inputs
text
Context
262,144 tokens
API price
No per-token rateA publisher-confirmed current standard hosted token price for Ling 3.0 Flash Fin has not been verified; an OpenRouter listing alone does not establish that rate.

What Ant Group says

  • Finance-enhanced continued-training checkpoint based on Ling 3.0 Flash.
  • Targets financial research, source-grounded retrieval, modeling, and spreadsheet workflows.

Limitations

  • The August 27 publisher launch date is verified; independent evaluation checkpoint matching remains incomplete.

Publisher statements, not AI IQ measurements.

About this score

How much of Ling 3.0 Flash Fin's IQ is measured, and how the rest is estimated.

8/ 39benchmark results
4/ 7dimensions measured
3estimated
0/5Abstract
0/5Math
3/6Academic
1/6Coding
0/5Computer Use
2/6Reliability
2/6Knowledge 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 and Computer Use have no published results yet, so their IQ is estimated from Ling 3.0 Flash Fin'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 Ling 3.0 Flash Fin

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

Ling 3.0 Flash FinComparison model
Browse all comparisons →

Frequently asked questions

Quick answers about Ling 3.0 Flash Fin, 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 Ling 3.0 Flash Fin?
Ling 3.0 Flash Fin has an estimated IQ of 102, which ranks #124 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 Ling 3.0 Flash Fin best at?
Among the dimensions with published benchmark results, Ling 3.0 Flash Fin ranks highest in Academic Reasoning (IQ 118, #117 of 223) and Reliability (IQ 111, #99 of 188). Its lowest-ranked measured dimension is Programmatic Reasoning (IQ 99, #140 of 187).
What is the EQ of Ling 3.0 Flash Fin?
Ling 3.0 Flash Fin 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 Ling 3.0 Flash Fin. AI IQ doesn't estimate EQ from other benchmarks, so the score appears once one of those results is published.
Which benchmarks has Ling 3.0 Flash Fin been tested on?
Ling 3.0 Flash Fin has published results on 8 of the 39 benchmarks behind AI IQ's Composite IQ, including AA Long Context Reasoning v1.1, CritPt, SciCode, AA Omniscience and AutomationBench-AA.
How much does Ling 3.0 Flash Fin cost?
AI IQ doesn't have a verified standard per-token API price for Ling 3.0 Flash Fin, so it can't calculate an effective cost. This is common for models offered only through subscriptions, restricted-access programs, or self-hosting.
Is Ling 3.0 Flash Fin open source?
Yes — Ling 3.0 Flash Fin is an open-weights model: its trained weights are published, so it can be downloaded and self-hosted.
Who developed Ling 3.0 Flash Fin?
Ling 3.0 Flash Fin was developed by Ant Group, based in China.
When was Ling 3.0 Flash Fin released?
Ling 3.0 Flash Fin was publicly released on August 27, 2026.
How large is the context window of Ling 3.0 Flash Fin?
Ling 3.0 Flash Fin supports a context window of 262,144 tokens — the amount of text it can consider in a single request.
How fast is Ling 3.0 Flash Fin?
In independent measurements, Ling 3.0 Flash Fin generates about 338 tokens per second, with a typical response time of about 9.2 seconds.