MODEL DIRECTORY
Follow the model.
See the evidence.
Benchmark results, evaluation settings and disclosed training methods—collected in one place for each model.
3,335 reported models and checkpoint variants. Scores keep their original units and sources.
636 models · page 9 of 14
OpenAI
GPT-5.4-high
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-5.4-medium
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-5.4-mini-2026-03-17†
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-5.4-mini-high
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-5.4-nano-2026-03-17†
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-5.4-nano-high
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-5.5 xhigh / Codex
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-5.5-2026-04-23-high-web-search†
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-5.5-2026-04-23†
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-5.5-medium
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-5.5-none
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-5.5-xhigh OpenAI
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-5.6 Luna (Terminus 2)
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-5.6 luna-max
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-5.6 Sol (Terminus 2)
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-5.6 Sol / Codex / third-party evaluation
1 benchmarks · 2 observations
Explore results →OpenAI
GPT-5.6 Sol / Exa
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-5.6 Sol / Keenable
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-5.6 Sol / Native
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-5.6 sol-max
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-5.6 sol-pro-max-web-search-code-execution-128k-cost-optimized†
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-5.6 sol-standard-max-web-search-code-execution-128k†
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-5.6 sol-standard-medium-web-search-code-execution†
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-5.6 Terra (Terminus 2)
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-5.6-terra-max
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-6 Astra (Codex) xHigh*
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-6 Astra (Terminus 2)
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-6 Astra / Codex
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-6 Astra + human collaborators
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-6 astra-max
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-6 Luna (Terminus 2)
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-6 Luna Decisions
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-6 luna-xhigh
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-6 Sol (Terminus 2)
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-6 sol-max
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-6.1 sol-max
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-f / Lean
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-f / Metamath
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-o3
1 benchmarks · 1 observations
Explore results →OpenAI
GPT-o4-mini
1 benchmarks · 1 observations
Explore results →OpenAI
gpt-realtime-1.5†
1 benchmarks · 4 observations
Explore results →OpenAI
gpt-realtime-2
1 benchmarks · 4 observations
Explore results →OpenAI
gpt-realtime-2025-08-28*
1 benchmarks · 4 observations
Explore results →OpenAI
gpt-realtime-mini-2025-12-15*
1 benchmarks · 4 observations
Explore results →OpenAI
GUA VL (GPT Eval)
1 benchmarks · 3 observations
Explore results →OpenAI
InstructBLIP (GPT Eval)
1 benchmarks · 3 observations
Explore results →NVIDIA
Llama 3.1-Nemotron-70B-Instruct-HF
1 benchmarks · 1 observations
Explore results →NVIDIA
Llama 3.1-Nemotron-70B-Reward
1 benchmarks · 1 observations
Explore results →