Sign Up

GPT-5 Nano

OpenAI
Announcement
GPT-5 Nano launch post cover

Cost/1M tokens

input
$0.05
cached
$0.005
output
$0.40
Context
400Ktokens
Released
1y ago

Benchmarks

DeepSWE is the primary coding benchmark where an exact model evaluation exists. Arena leaderboards and other sourced evaluations appear below; missing scores are not zero.

How it ranks

Against its predecessor, similarly priced models and the models ranked around it.

DeepSWE 1.1

Best pass@1 per model · Higher is better

Not evaluated in DeepSWE 1.1

Arena Text

Best Arena rating per model · Higher is better

Arena #241

  1. 1454
    Hy3Similar price
  2. 1437
    Gemma 4 26B A4BSimilar price
  3. 1364
    GLM 4.7 FlashSimilar price
  4. 1347
    Qwen3 32BSimilar price
  5. 1343
    Llama 3.3 Nemotron Super 49B v1.5Nearby rank
  6. 1342
    Gemma 3 12BNearby rank
  7. 1338
    GPT-5 Nano
  8. 1336
    GPT-4o (2024-08-06)Nearby rank
  9. 1336
    QwQ 32BNearby rank
  10. 1322
    GPT-4.1 NanoPredecessor

+15 vs GPT-4.1 Nano (predecessor) · Ahead of 0 of 4 similarly priced models

Arena WebDev

Best Arena rating per model · Higher is better

Not ranked on Arena WebDev yet

Arena Vision

Best Arena rating per model · Higher is better

Arena #120

  1. 1248
    MiMo-V2.6-FlashSimilar price
  2. 1242
    Gemma 4 26B A4BSimilar price
  3. 1155
    Mistral Medium 3Nearby rank
  4. 1153
    GLM 4.5VNearby rank
  5. 1147
    GPT-5 Nano
  6. 1145
    Llama 4 MaverickNearby rank
  7. 1140
    Mistral Small 3.2Nearby rank
  8. 1128
    Llama 4 ScoutSimilar price
  9. 1089
    GPT-4.1 NanoPredecessor
  10. 1019
    Nova Lite 1.0Similar price

+59 vs GPT-4.1 Nano (predecessor) · Ahead of 1 of 4 similarly priced models

Arena Search

Best Arena rating per model · Higher is better

Not ranked on Arena Search yet

Arena Document

Best Arena rating per model · Higher is better

Not ranked on Arena Document yet

Agent Arena

Best IPS score per model · Higher is better

Not ranked on Agent Arena yet

ARC-AGI-2

Best score per model · Higher is better

#70 of 85

  1. 78.1%
    DeepSeek V4.1 Flash 1017Similar price
  2. 61.4%
    DeepSeek V4 Flash 0731Similar price
  3. 61.3%
    Claude Haiku 5.5Similar price
  4. 4.0%
    Claude Haiku 4.5Nearby rank
  5. 4.0%
    DeepSeek V3.2Nearby rank
  6. 2.6%
    GPT-5 Nano
  7. 2.5%
    Gemini 2.5 Flash Preview 05-20Nearby rank
  8. 1.3%
    Gemini 3 Flash PreviewNearby rank
  9. 0.0%
    Llama 4 ScoutSimilar price
  10. 0.0%
    GPT-4.1 NanoPredecessor

+2.6 pts vs GPT-4.1 Nano (predecessor) · Ahead of 1 of 4 similarly priced models

FrontierMath-2025-02-28-Private

Best score per model · Higher is better

#34 of 58

  1. 10.3%
    Qwen3.6 FlashNearby rank
  2. 8.5%
    Qwen3 235B A22B Thinking 2507Nearby rank
  3. 8.3%
    GPT-5 Nano
  4. 7.2%
    Claude Opus 4.1Nearby rank
  5. 6.2%
    Qwen3.5-FlashSimilar price
  6. 1.0%
    GPT-4.1 NanoPredecessor
  7. 0.0%
    Llama 4 ScoutSimilar price

+7.2 pts vs GPT-4.1 Nano (predecessor) · Ahead of 2 of 2 similarly priced models

FrontierMath-Tier-4-2025-07-01-Private

Best score per model · Higher is better

#32 of 46

  1. 2.1%
    Claude Haiku 4.5Nearby rank
  2. 2.1%
    Gemini 2.5 Pro Preview 06-05Nearby rank
  3. 2.1%
    GPT-5 Nano
  4. 2.1%
    o3Nearby rank
  5. 2.1%
    GPT-5.4 MiniNearby rank
  6. 0.0%
    Qwen3.5-FlashSimilar price

Ahead of 1 of 1 similarly priced model

FrontierMath-Tier-4-v2-Private

Best score per model · Higher is better

#64 of 72

  1. 56.1%
    GPT-6 LunaSimilar price
  2. 46.3%
    Claude Haiku 5.5Similar price
  3. 46.3%
    DeepSeek V4.1 Flash 1017Similar price
  4. 24.4%
    DeepSeek V4 Flash 0731Similar price
  5. 2.4%
    Claude Sonnet 4.5Nearby rank
  6. 2.4%
    DeepSeek V4 Pro 0423Nearby rank
  7. 2.4%
    GPT-5 Nano
  8. 2.4%
    GPT-5.5 InstantNearby rank
  9. 0.0%
    Gemini 2.5 ProNearby rank

Ahead of 0 of 4 similarly priced models

FrontierMath-Tiers-1-3-v2-Private

Best score per model · Higher is better

#78 of 87

  1. 75.1%
    Claude Haiku 5.5Similar price
  2. 71.2%
    DeepSeek V4.1 Flash 1017Similar price
  3. 57.5%
    DeepSeek V4 Flash 0731Similar price
  4. 22.5%
    Qwen3.6 FlashNearby rank
  5. 20.4%
    Qwen3.6 35B A3BNearby rank
  6. 20.0%
    GPT-5 Nano
  7. 19.3%
    Qwen3.7 FlashNearby rank
  8. 18.9%
    Qwen3 MaxNearby rank
  9. 18.2%
    Qwen3.5-FlashSimilar price

Ahead of 1 of 4 similarly priced models

GPQA diamond

Best score per model · Higher is better

#111 of 161

  1. 73.2%
    Gemma 4 26B A4BSimilar price
  2. 71.2%
    DeepSeek V3Nearby rank
  3. 70.7%
    Qwen3 235B-A22BNearby rank
  4. 69.4%
    GPT-5 Nano
  5. 67.6%
    DeepSeek V3 0324Nearby rank
  6. 66.9%
    GPT-4.1Nearby rank
  7. 65.7%
    Qwen3 32BSimilar price
  8. 60.5%
    GLM 4.7 FlashSimilar price
  9. 51.8%
    Llama 4 ScoutSimilar price
  10. 48.9%
    GPT-4.1 NanoPredecessor

+20.5 pts vs GPT-4.1 Nano (predecessor) · Ahead of 3 of 4 similarly priced models

OTIS Mock AIME 2024-2025

Best score per model · Higher is better

#84 of 158

  1. 83.1%
    Kimi K2 Thinking TurboNearby rank
  2. 82.2%
    Gemma 4 26B A4BSimilar price
  3. 81.1%
    GPT-5 Nano
  4. 80.0%
    Gemini 3.1 Flash LiteNearby rank
  5. 80.0%
    GLM 5Nearby rank
  6. 66.9%
    Qwen3 32BSimilar price
  7. 58.3%
    GLM 4.7 FlashSimilar price
  8. 28.9%
    GPT-4.1 NanoPredecessor
  9. 7.8%
    Llama 4 ScoutSimilar price

+52.2 pts vs GPT-4.1 Nano (predecessor) · Ahead of 3 of 4 similarly priced models

SimpleQA Verified

Best score per model · Higher is better

#83 of 87

  1. 33.6%
    DeepSeek V4 Flash 0731Similar price
  2. 23.8%
    Claude Haiku 5.5Similar price
  3. 20.3%
    Qwen3.5-FlashSimilar price
  4. 13.2%
    Claude Haiku 4.5Nearby rank
  5. 12.7%
    GPT-4.1 MiniNearby rank
  6. 11.7%
    GPT-5 Nano
  7. 11.7%
    GPT-5.4 NanoNearby rank
  8. 10.4%
    Gemma 4 31BSimilar price
  9. 6.0%
    GPT-4.1 NanoPredecessor

+5.7 pts vs GPT-4.1 Nano (predecessor) · Ahead of 1 of 4 similarly priced models

Terminal Bench

Best score per model · Higher is better

#45 of 49

  1. 24.6%
    Qwen3.6 35B A3BNearby rank
  2. 24.5%
    GLM 4.6Nearby rank
  3. 21.8%
    GPT-5 Nano
  4. 18.7%
    gpt-oss-120bSimilar price
  5. 17.1%
    Gemini 2.5 FlashNearby rank
  6. 9.2%
    Qwen3.5-9BSimilar price

Ahead of 2 of 2 similarly priced models

Not evaluated in DeepSWE 1.1 as of 2026-09-22. Results are only shown for exact model versions in the official dataset. Official leaderboard.

Reasoning effort

ARC-AGI-2

Score by reasoning effort · Higher is better

FrontierMath-2025-02-28-Private

Score by reasoning effort · Higher is better

FrontierMath-Tier-4-2025-07-01-Private

Score by reasoning effort · Higher is better

FrontierMath-Tiers-1-3-v2-Private

Score by reasoning effort · Higher is better

GPQA diamond

Score by reasoning effort · Higher is better

OTIS Mock AIME 2024-2025

Score by reasoning effort · Higher is better

Terminal Bench

Score by reasoning effort · Higher is better

All published scores

Benchmark compilation: Epoch AI, Capabilities & benchmarking (CC BY attribution). Scores are converted to percentages and model names matched to the Vertical Tabs catalog. Individual evaluations are credited to their original sources. External results may carry separate licences and evaluation configurations; do not interpret scores from different harnesses as directly comparable.

Arena ratings: Arena (lmarena-ai/leaderboard-dataset), licensed CC BY 4.0. Ratings come from pairwise human votes and are only comparable within one leaderboard; the interval is Arena's 95% confidence interval. Model names are matched to the Vertical Tabs catalog.

To give you the best experience, we use cookies for personalization, insights, and performance. Learn more.