The Latent
EN▾
EnglishEspañol中文PortuguêsFrançaisالعربية日本語한국어
Sign Up
NEWSDATARESEARCHPODCASTNEWSLETTERPRESSPARTNER WITH US →
Data/Capability

Capability

Historical Inference Speeds

Historical Inference Speeds. Demo series compiled by The Latent research desk.

Historical Inference Speeds

  • Groq
  • Google
  • OpenAI
  • Anthropic
Historical Inference SpeedsDemo preview through July 2026 (tokens per second).3K2K1K0Jul '23Jan '24Jul '24Jan '25Jul '25Jan '26Jul '26Groq: 2.3K on Jul '26. Google: 841.6 on Jul '26. OpenAI: 621.2 on Jul '26. Anthropic: 561.1 on Jul '26
SOURCE: The LatentUPDATED: Jul 1, 2026
ZOOMALLYTD12M3M1M
CSV
View data
DateGroqGoogleOpenAIAnthropic
2023-08-011115.23407.18300.45271.29
2023-09-011228.16448.2330.55298.32
2023-10-011273.1465.19343.52310.42
2023-11-011213.57443.37327.36295.78
2023-12-011250.45456.76337.18304.6
2024-01-011204.64440.25325.17293.9
2024-02-011216.69444.04327.51295.61
2024-03-011358.21495.8365.76330.19
2024-04-011372.07501.13369.91334.13
2024-05-011436.66524.49386.97349.38
2024-06-011468.69536.77396.46358.33
2024-07-011358.9496.6366.76331.48
2024-08-011408.46514.22379.42342.58
2024-09-011443.78527.39389.33351.72
2024-10-011473.63537.82396.68358.05
2024-11-011648.98601.92444.04400.85
2024-12-011643.41600.54443.51400.82
2025-01-011619.85591.61436.68394.43
2025-02-011656.27605.22446.94403.89
2025-03-011554.95568.17419.57379.15
2025-04-011643.93599.88442.4399.25
2025-05-011780.26650.09479.75433.26
2025-06-011794.15655.08483.38436.49
2025-07-011935.11706.54521.34470.75
2025-08-011899.53694.38512.99463.77
2025-09-011796.83656.37484.56437.76
2025-10-011885.67688.65508.26459.05
2025-11-011862.12680.19502.13453.61
2025-12-011979.8722.35532.64480.63
2026-01-012181.21796.5587.81530.84
2026-02-012129.91778.16574.57519.16
2026-03-012182.04796.88588.15531.2
2026-04-012156.32788.2582.27526.37
2026-05-012036.2743.7548.96495.88
2026-06-012218.42809.66597.19539.03
2026-07-012304.47841.64621.22561.11

API access · Pro coming soon

Demo preview through July 2026 (tokens per second).

Download CSVPro API coming soon

Methodology

Synthetic demo series generated to preview the data product layout.

Related charts

  • Historical Token Throughput per GPU
  • Historical Token Efficiency Across Frontier Models
  • Epoch Capabilities Index (ECI)

The Latent

AI industry news. A sister publication to The Block.

Editorial

  • Standards
  • Corrections
  • Commercial policy
  • Contact

Company

  • About
  • Privacy Policy
  • Terms of Service
  • Security
  • The Block