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/ByteDance
ByteDance

ByteDance · Seed-OSS

Seed-OSS-36B-Instruct

Open weightsSupersededApache-2.0

ByteDance's only open-weight dense text model (36B, 512K context); marked deprecated on Artificial Analysis and not refreshed since 2025.

AA Intelligence
12.1#88 of 142
Price
Self-host
Output speed
41tok/s
Context window
512K tokens

Composite indexes

  • Artificial Analysis Intelligence Index#88 of 142
    12independentv4.3.2 (AA estimate; independent evaluation forthcoming)Artificial Analysis: Seed-OSS-36B-Instruct ↗

    Leader: Claude Opus 5.5 at 58

Coding

  • LiveCodeBench#15 of 56
    76.5%independentAAArtificial Analysis: Seed-OSS-36B-Instruct ↗

    Leader: Fugu at 92.9%

Agents and tool use

  • τ²-Bench#46 of 88
    49.4%independentAA, telecomArtificial Analysis: Seed-OSS-36B-Instruct ↗

    Leader: GLM-5.2 at 99.1%

Reasoning and knowledge

  • Humanity’s Last Exam#105 of 147
    9.9%independentAA, no toolsArtificial Analysis: Seed-OSS-36B-Instruct ↗

    Leader: Claude Opus 5.5 at 61.4%

  • GPQA Diamond#97 of 142
    72.6%independentAAArtificial Analysis: Seed-OSS-36B-Instruct ↗

    Leader: GPT-6 Astra at 96.1%

  • CritPt#99 of 139
    0%independentAAArtificial Analysis: Seed-OSS-36B-Instruct ↗

    Leader: GPT-5.6 Sol at 32.3%

Vision and long context

  • AA Long Context Reasoning#83 of 139
    61.3%independentAAArtificial Analysis: Seed-OSS-36B-Instruct ↗

    Leader: Kimi K3 at 88.7%

Other published results

ResultScoreReported bySource
aime-2025 · AA, AIME 202584.7%independentArtificial Analysis: Seed-OSS-36B-Instruct ↗
Terminal-Bench Hard (AA) · AA6.8%independentArtificial Analysis: Seed-OSS-36B-Instruct ↗
IFBench (AA) · AA41.9%independentArtificial Analysis: Seed-OSS-36B-Instruct ↗
AA-Omniscience Index · AA-52.5 index (-100..100)independentArtificial Analysis: Seed-OSS-36B-Instruct ↗

Specs

Released
August 20, 2025
Parameters
36.2B
Size class
10 to 40B
Input
text
Output
text
Context
512K tokens
Max output
Not published
First token
2.98s
  • Speed: Artificial Analysis: Seed-OSS-36B-Instruct (median across providers) ↗

Will it fit your budget?

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

Plan a stack →

Compare with

  • InceptionMercury 2.512.3
  • MistralMagistral Medium 1.211.8
  • PerplexitySonar Reasoning Pro (Sonar API)11.8
  • NVIDIANemotron Cascade 2 30B A3B11.7