modelgrep

Fastest OpenAI Models

Quick answer · Updated June 2026

The fastest OpenAI model is gpt-oss-safeguard-20b at 570 output tokens per second. gpt-oss-120b (free) (547 t/s) and gpt-oss-120b (547 t/s) round out the top three.

570 t/sSpeed
$0.075Input /M
131KContext

AI models ranked by output speed (tokens per second, p50). The fastest large language models for low-latency and high-throughput applications.

  1. 1O
    gpt-oss-safeguard-20b
    ReasoningToolsJSON$0.075/M · 229ms ttft · 131K ctx
    570 t/s
    Speed
  2. 2O
    gpt-oss-120b:free
    ReasoningTools33.3 intel · Free/M · 177ms ttft
    547 t/s
    Speed
  3. 3O
    gpt-oss-120b
    ReasoningToolsJSON33.3 intel · $0.039/M · 177ms ttft
    547 t/s
    Speed
  4. 4O
    gpt-oss-20b:free
    ReasoningToolsJSON24.5 intel · Free/M · 247ms ttft
    391 t/s
    Speed
  5. 5O
    gpt-oss-20b
    ReasoningToolsJSON24.5 intel · $0.029/M · 247ms ttft
    391 t/s
    Speed
  6. 6O
    o3-mini
    ReasoningToolsJSON$1.10/M · 2.2s ttft · 200K ctx
    144 t/s
    Speed
  7. 7O
    o4-mini-deep-research
    ReasoningToolsJSON+1$2.00/M · 2.4s ttft · 200K ctx
    130 t/s
    Speed
  8. 8O
    gpt-5.1-codex-mini
    ReasoningToolsJSON+138.6 intel · $0.250/M · 1.7s ttft
    127 t/s
    Speed
  9. 9O
    o4-mini-high
    ReasoningToolsJSON+1$1.10/M · 4.3s ttft · 200K ctx
    124 t/s
    Speed
  10. 10O
    o4-mini
    ReasoningToolsJSON+133.1 intel · $1.10/M · 4.5s ttft
    122 t/s
    Speed
  11. 11O
    gpt-5-image-mini
    ReasoningJSONVision+1$2.50/M · 6.1s ttft · 400K ctx
    110 t/s
    Speed
  12. 12O
    gpt-4.1-nano
    ToolsJSONVision13.0 intel · $0.100/M · 649ms ttft
    91 t/s
    Speed
  13. 13O
    gpt-4o-2024-08-06
    ToolsJSONVision18.6 intel · $2.50/M · 722ms ttft
    84 t/s
    Speed
  14. 14O
    gpt-5.4-mini
    ReasoningToolsJSON+123.3 intel · $0.750/M · 728ms ttft
    83 t/s
    Speed
  15. 15O
    gpt-5-image
    ReasoningJSONVision+1$10.00/M · 7.9s ttft · 400K ctx
    83 t/s
    Speed
  16. 16O
    gpt-5.1-codex
    ReasoningToolsJSON+143.1 intel · $1.25/M · 2.8s ttft
    79 t/s
    Speed
  17. 17O
    gpt-5-mini
    ReasoningToolsJSON+141.2 intel · $0.250/M · 729ms ttft
    70 t/s
    Speed
  18. 18O
    gpt-5.1-codex-max
    ReasoningToolsJSON+1$1.25/M · 1.6s ttft · 400K ctx
    67 t/s
    Speed
  19. 19O
    gpt-5.1
    ReasoningToolsJSON+147.7 intel · $1.25/M · 1.0s ttft
    67 t/s
    Speed
  20. 20O
    gpt-chat-latest
    ToolsJSONVision$5.00/M · 1.3s ttft · 400K ctx
    66 t/s
    Speed
  21. 21O
    gpt-5.2-codex
    ReasoningToolsJSON+149.0 intel · $1.75/M · 1.5s ttft
    63 t/s
    Speed
  22. 22O
    gpt-5.4
    ReasoningToolsJSON+156.8 intel · $2.50/M · 982ms ttft
    62 t/s
    Speed
  23. 23O
    gpt-5.1-chat
    ToolsJSONVision$1.25/M · 1.1s ttft · 128K ctx
    61 t/s
    Speed
  24. 24O
    gpt-5.4-nano
    ReasoningToolsJSON+144.0 intel · $0.200/M · 552ms ttft
    60 t/s
    Speed
  25. 25O
    gpt-audio-mini
    ToolsJSONAudio$0.600/M · 459ms ttft · 128K ctx
    60 t/s
    Speed

Frequently asked

What is the fastest OpenAI model?

The fastest OpenAI model is gpt-oss-safeguard-20b at 570 output tokens per second. gpt-oss-120b (free) (547 t/s) and gpt-oss-120b (547 t/s) round out the top three.

What's a good alternative to gpt-oss-safeguard-20b?

gpt-oss-120b (free) (547 t/s) is the closest alternative on this metric, followed by gpt-oss-120b (547 t/s). See the full ranking above for the tradeoffs.

How many OpenAI models are there?

modelgrep tracks 62 OpenAI models with live benchmarks, speed, latency and per-provider pricing, led on intelligence by GPT-5.4. 25 of them qualify for this ranking.

More OpenAI rankings

All rankings