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/China Mobile
China Mobile

China Mobile · JT (Jiutian)

JT-4.1 Flash 236B A21B

Closed weightsGenerally availableproprietary

China Mobile's Jiutian 236B/21B-active MoE model; listed on Artificial Analysis with an estimated index score only.

AA Intelligence
27.3#44 of 142
Price
—
Output speed
—
Context window
256K tokens

Composite indexes

  • Artificial Analysis Intelligence Index#44 of 142
    27independentv4.3.2 (AA estimate; independent evaluation forthcoming)Artificial Analysis: JT-4.1 Flash 236B A21B ↗

    Leader: Claude Opus 5.5 at 58

Coding

  • Terminal-Bench 2.x#50 of 103
    59.6%independentAA, Terminal-Bench 2.1Artificial Analysis: JT-4.1 Flash 236B A21B ↗

    Leader: Claude Fable 5.1 at 91.4%

Agents and tool use

  • GDPval-AA#45 of 103
    1114independentGDPval-AA v2.1Artificial Analysis: JT-4.1 Flash 236B A21B ↗

    Leader: Claude Opus 5.5 at 1846

Reasoning and knowledge

  • Humanity’s Last Exam#77 of 147
    17.8%independentAA, no toolsArtificial Analysis: JT-4.1 Flash 236B A21B ↗

    Leader: Claude Opus 5.5 at 61.4%

  • GPQA Diamond#60 of 142
    84.5%independentAAArtificial Analysis: JT-4.1 Flash 236B A21B ↗

    Leader: GPT-6 Astra at 96.1%

  • CritPt#103 of 139
    0%independentAAArtificial Analysis: JT-4.1 Flash 236B A21B ↗

    Leader: GPT-5.6 Sol at 32.3%

Vision and long context

  • AA Long Context Reasoning#73 of 139
    71.3%independentAAArtificial Analysis: JT-4.1 Flash 236B A21B ↗

    Leader: Kimi K3 at 88.7%

  • MMMU-Pro#52 of 72
    64.1%independentAAArtificial Analysis: JT-4.1 Flash 236B A21B ↗

    Leader: Claude Opus 5.5 at 87.7%

Other published results

ResultScoreReported bySource
AA-Omniscience Index · AA-8.8 index (-100..100)independentArtificial Analysis: JT-4.1 Flash 236B A21B ↗

Specs

Released
July 9, 2026
Parameters
236B total, 21B active
Size class
100 to 500B
Input
text, image
Output
text
Context
256K tokens
Max output
Not published

    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 Build 0.127.2
    • Moonshot AIKimi K2.627
    • Thinking Machines LabInkling-Small27.8
    • Multiverse ComputingQuasar 438B26.7