Skip to content
#22B+Jul 30, 2026Evidence B

Ant Group Ling 2

Ant Group · Undisclosed (MoE, finance-optimised)

The finance-scenario model with Chinese 92

Ant’s general model, with Chinese context 92 — eighth on this site, behind Doubao’s 97, the three 95s at Qwen3.8-Max/GLM-5/GLM-5.3, Qwen3.5’s 94 and the two 93s at Kimi K3/ERNIE. Composite 80.8, ranked 22nd: not high, but its accumulation on financial instruments, compliance text and payment documents is exclusive. Grade B: three months old with limited third-party reproduction.

TextImageNative

Last updated Oct 18, 2026Editorial scoring

Bottom line

For finance and payment scenarios — instruments, compliance, risk documents — it belongs on the shortlist: Chinese 92 plus scenario corpora is something score peers cannot offer. For general multimodal, skip it: video 60, generation 68, and an ecosystem bound to Ant’s stack. It is also an escaper in the 9-dimension dominance analysis: dominated on price by cheaper models, but with no rival on the Chinese dimension.

Eight-dimension breakdown

editorial mapping · 0–100
Perception90Document88Chart QA82Video60Agent80Generation68Chinese92Ecosystem788D
  • Ant Group Ling 280.8
Visual perception & grounding20%
90
Document parsing & Chinese OCR18%
88
Chart & document reasoning14%
82
Video & long-horizon reasoning14%
60
GUI agents & tool use14%
80
Generation & editing6%
68
Chinese context & culture6%
92
Ecosystem, cost & availability8%
78

Weights: perception 20 / documents 18 / charts 14 / video 14 / agents 14 / generation 6 / Chinese 6 / ecosystem 8. Definitions and sources on the methodology page.

Public benchmarks

as reported, not converted
Benchmark / metricValueNote
中文金融文本评测91.5Vendor figure
输出单价¥5Official API

Public scores come from different versions and prompts and are not interchangeable with our composite; they are one evidence input.

Scenario mapping

five stress tests, one scale
ScenarioScoreObservation
Chinese receipt OCR88 · Boosted by financial-instrument scenarios
Chart data extraction80 · Reliable on standard charts
30-min video QA56 · Summary level only
Mobile GUI control74 · Not trained for GUI
Chinese cultural context92 · Same as our Chinese dimension

Strengths and limitations

Strengths

  • Chinese 92: exclusive register priors on financial text, compliance language and policy documents.
  • Deep scenario corpora on financial instruments and compliance documents.
  • 256K context at ¥5 — a rare strong-Chinese option at mid price.

Limitations

  • Video 60, generation 68: general multimodal capability is priced in as skewed.
  • Ecosystem bound to Ant’s stack: read the terms on payment, chain and cloud restrictions carefully.
  • Grade B: three months old, third-party reproduction still limited.

Price and real cost

List price

¥5/M tokens

≈ ¥5 / 1M output tokens. For 1,000 images at 1,024px, ~1,100 image tokens each and 300 output tokens, the monthly cost is about ¥3.0 (assuming input at 25% of output price).

Use your volume

About Ant Group Ling 2

Is Ant Group Ling 2 open source?

Licence: Closed commercial. No public weights; access is via vendor API or an on-prem licence.

How much does Ant Group Ling 2 cost?

≈ ¥5 / 1M output tokens. It ranks #22 with a composite of 80.8. Real bills depend on resolution, tiling, caching and volume discounts.

What is Ant Group Ling 2 best for?

Financial instrument and compliance document processing; Chinese long-form text and policy interpretation; Payment and risk-control QA。

Why does Ant Group Ling 2 score 80.8?

The composite weights eight dimensions: perception 20%, documents 18%, charts 14%, video 14%, agents 14%, generation 6%, Chinese 6%, ecosystem 8%. Its strongest dimension is Chinese context & culture; its weakest is Video & long-horizon reasoning.

Personal ranking · Not an official score

Every ranking, score and comparison on this page is an independent personal ranking, not an official score, and represents no vendor, institution or regulator. Scores are a subjective mapping of public evidence under our own weights — change the weights and the order changes. Disagree, or spotted a factual error? Contact the site owner: jizhi2024@foxmail.com — we check each report and log corrections in the changelog.