Skip to contentT/TTokens or Towers

Compare

LeaderboardEvery tracked model, ranked and priced side by sideBenchmarksWhat each test measures and who leads itPricesAPI, media, and hardware rates with a cost calculator

Plan

Stack plannerSix questions to a priced local and hosted stackShortlistOur picks for subscriptions, APIs, and hardwareEnterpriseSeat plans against self-hosting for a whole team

Build

HardwareGPUs and workstations, and which models they fitHostsWhere to rent open-weight inference, and for how muchHarnessesCoding agents and the models they drive

Lab

FindingsResults from our own evaluationsScanAudit a repo’s AI usage in the browserMethodWhere the numbers come from and how we check them
Data checked September 25, 2026
Data checked September 25, 2026
Plan a stack
T/TTokens or Towers

An independent guide to AI models, what they cost, and the hardware to run them. No account, no API key, no sponsors.

Compare

  • Leaderboard
  • Benchmarks
  • Prices

Plan

  • Stack planner
  • Shortlist
  • Enterprise

Build

  • Hardware
  • Hosts
  • Harnesses

Lab

  • Findings
  • Scan
  • Method
Data checked September 25, 2026 · Every number links to its sourceRSS · Source on GitHub
Leaderboard/Ant Group
Ant Group

Ant Group · Ling 3.0

Ling-3.0-flash-VL

Open weightsGenerally availableNewMIT

Ant Group's open-weight (MIT) 124B/5.5B-active natively multimodal MoE reasoning model with 256K context, very cheap to run.

AA Intelligence
24.6#58 of 142
Price
Self-host
Output speed
146tok/s
Context window
262K tokens

Composite indexes

  • Artificial Analysis Intelligence Index#58 of 142
    25independentv4.3.2Artificial Analysis: Ling-3.0-flash-VL ↗

    Leader: Claude Opus 5.5 at 58

Coding

  • Terminal-Bench 4.0#73 of 88
    0%independentAAArtificial Analysis: Ling-3.0-flash-VL ↗

    Leader: Claude Mythos 5.1 at 60.9%

  • Terminal-Bench 2.x#44 of 103
    64.4%independentAA, Terminal-Bench 2.1Artificial Analysis: Ling-3.0-flash-VL ↗

    Leader: Claude Fable 5.1 at 91.4%

  • SciCode#59 of 94
    44.2%independentAAArtificial Analysis: Ling-3.0-flash-VL ↗

    Leader: Claude Opus 5.5 at 66.9%

Agents and tool use

  • GDPval-AA#43 of 103
    1150independentGDPval-AA v2.1Artificial Analysis: Ling-3.0-flash-VL ↗

    Leader: Claude Opus 5.5 at 1846

Reasoning and knowledge

  • Humanity’s Last Exam#69 of 147
    22%independentAA, no toolsArtificial Analysis: Ling-3.0-flash-VL ↗

    Leader: Claude Opus 5.5 at 61.4%

  • GPQA Diamond#54 of 142
    86.2%independentAAArtificial Analysis: Ling-3.0-flash-VL ↗

    Leader: GPT-6 Astra at 96.1%

  • CritPt#64 of 139
    2%independentAAArtificial Analysis: Ling-3.0-flash-VL ↗

    Leader: GPT-5.6 Sol at 32.3%

Vision and long context

  • AA Long Context Reasoning#47 of 139
    78.3%independentAAArtificial Analysis: Ling-3.0-flash-VL ↗

    Leader: Kimi K3 at 88.7%

  • MMMU-Pro#22 of 72
    79%independentAAArtificial Analysis: Ling-3.0-flash-VL ↗

    Leader: Claude Opus 5.5 at 87.7%

Other published results

ResultScoreReported bySource
AA-Omniscience Index · AA-4.5 index (-100..100)independentArtificial Analysis: Ling-3.0-flash-VL ↗
AA Intelligence Index v4.1.1 (vendor-quoted) · old index version; not comparable to v4.3.242 indexvendorHugging Face: inclusionAI/Ling-3.0-flash-VL ↗

Specs

Released
September 10, 2026
Parameters
124B total, 5.5B active
Size class
100 to 500B
Input
text, image, video
Output
text
Context
262K tokens
Max output
Not published
First token
1.91s
  • Speed: Artificial Analysis: Ling-3.0-flash-VL (InclusionAI first-party endpoint) ↗

Will it fit your budget?

The planner prices this model against local hardware and other APIs for your workload.

Plan a stack →

Compare with

  • xAIGrok 4.324.9
  • Ant GroupLing 3.0 Flash24.9
  • AlibabaQwen3.7-Plus25
  • Thinking Machines LabInkling25