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/AI21 Labs
AI21 Labs

AI21 Labs · Jamba

Jamba Reasoning 3B

Open weightsGenerally availableApache 2.0

AI21's 3B open reasoning model with long (262K) context for on-device use.

AA Intelligence
5.7#134 of 142
Price
Self-host
Output speed
—
Context window
262K tokens

Composite indexes

  • Artificial Analysis Intelligence Index#134 of 142
    6independentIntelligence Index v4.3; reasoning; estimatedArtificial Analysis: Jamba Reasoning 3B ↗

    Leader: Claude Opus 5.5 at 58

Coding

  • LiveCodeBench#50 of 56
    21%independentreasoningArtificial Analysis: Jamba Reasoning 3B ↗

    Leader: Fugu at 92.9%

Agents and tool use

  • τ²-Bench#72 of 88
    15.8%independenttelecom; reasoningArtificial Analysis: Jamba Reasoning 3B ↗

    Leader: GLM-5.2 at 99.1%

Reasoning and knowledge

  • GPQA Diamond#137 of 142
    33.3%independentreasoningArtificial Analysis: Jamba Reasoning 3B ↗

    Leader: GPT-6 Astra at 96.1%

  • Humanity’s Last Exam#143 of 147
    3.8%independentno tools; reasoningArtificial Analysis: Jamba Reasoning 3B ↗

    Leader: Claude Opus 5.5 at 61.4%

  • CritPt#133 of 139
    0%independentreasoningArtificial Analysis: Jamba Reasoning 3B ↗

    Leader: GPT-5.6 Sol at 32.3%

Vision and long context

  • AA Long Context Reasoning#126 of 139
    6.3%independentreasoningArtificial Analysis: Jamba Reasoning 3B ↗

    Leader: Kimi K3 at 88.7%

Other published results

ResultScoreReported bySource
aime-2025 · AIME 2025; reasoning10.7%independentArtificial Analysis: Jamba Reasoning 3B ↗
Terminal-Bench Hard (AA) · reasoning0.8%independentArtificial Analysis: Jamba Reasoning 3B ↗
IFBench · reasoning52.4%independentArtificial Analysis: Jamba Reasoning 3B ↗
AA-Omniscience Index · reasoning-57.7 indexindependentArtificial Analysis: Jamba Reasoning 3B ↗

Specs

Released
October 8, 2025
Parameters
3B
Size class
Under 10B
Input
text
Output
text
Context
262K 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

    • MicrosoftPhi-4 Multimodal Instruct5.8
    • Ai2Olmo 3 7B Think5.6
    • Reka AIReka Flash 35.6
    • MistralMinistral 3 8B5.5