ResearchLLM Verifications Leaderboard

Data as of · 78 models, 80 questions · reviewed by Suhas Bhairav · methodology

Provider

OpenAI AI models

13 models tested, compared with 65 models from other labs.

Best verification score82.5GPT-5.3-Codex · #3 of 78Best output speed163o3 · #14 of 78Best time to first token0.5 sgpt-oss-120b · #20 of 78Best cost per question$0.000066gpt-oss-20b · #2 of 78
OpenAI#3GPT-5.3-Codex82.5% verification score95 tok/s1.4 s first token$0.01 / questionOpenAI#12GPT-6 Astra76.2% verification score49 tok/s1.3 s first token$0.013 / questionOpenAI#12GPT-6.1 Terra Pro76.2% verification score— tok/s8.5 s first token$0.005 / questionOpenAI#12GPT-6.1 Sol76.2% verification score41 tok/s2.1 s first token$0.0029 / questionOpenAI#15GPT-5.6 Terra75.0% verification score101 tok/s1.1 s first token$0.0063 / questionOpenAI#18GPT-6.1 Terra73.8% verification score122 tok/s2.4 s first token$0.0013 / questionOpenAI#18GPT-6 Sol73.8% verification score92 tok/s0.9 s first token$0.0024 / questionOpenAI#21o372.5% verification score163 tok/s1.2 s first token$0.0088 / questionOpenAI#21GPT-5.472.5% verification score55 tok/s1.7 s first token$0.011 / questionOpenAI#21GPT-6 Luna72.5% verification score137 tok/s1.2 s first token$0.00025 / questionOpenAI#41gpt-oss-120b58.8% verification score82 tok/s0.5 s first token$0.00013 / questionOpenAI#46gpt-oss-20b51.2% verification score103 tok/s0.7 s first token$0.000066 / question

Score vs. cost per question

OpenAI models in color, every other model in grey

Most attractive quadrant0255075100$0.00001$0.0001$0.001$0.01$0.10Cost per question (USD, log) →Verification score (%) →Qwen3.8 2.4T A95BCommand AReka Flash 3Gemma 4 26B A4BGemma 4 31BQwen3.5 397B A17BHermes 4 405BGPT-6 AstraGrok 4.7o3gpt-oss-20bMinistral 3 3B 2512Nova Micro 1.0Phi 4Granite 4.2 8BHy3GLM 5.3 FlashSolar Pro 4Trinity Large ThinkingMiniMax M3Mistral Large 4Claude Opus 5.5GPT-6.1 SolGemini 4 ArgonGPT-6 Lunagpt-oss-120bLlama 4 Maverick

Output speed vs. cost per question

OpenAI models in color, every other model in grey

Most attractive quadrant0200400600800$0.00001$0.0001$0.001$0.01$0.10Cost per question (USD, log) →Output speed (tok/s) →Nemotron 3.5 Super VLQwen3.5 397B A17BMuse WonderGrok 4.7o3gpt-oss-20bMinistral 3 3B 2512Nova Micro 1.0Trinity Large ThinkingInkling SmallGemini 3.5 Flash LiteMuse Spark OpenGPT-6 Luna

Verification score

OpenAI models highlighted against every model in this run · higher is better

  1. 1Claude Opus 5.585.0
  2. 2Gemini 4 Argon83.8
  3. 3GPT-5.3-Codex82.5
  4. 3Gemini 3.8 Flash82.5
  5. 5Claude Fable 5.181.2
  6. 6Grok 4.780.0
  7. 7Muse Wonder78.8
  8. 7Claude Sonnet 5.578.8
  9. 9Gemini 3.1 Pro Preview77.5
  10. 9Muse Spark Open77.5
  11. 12GPT-6 Astra76.2
  12. 12GPT-6.1 Terra Pro76.2
  13. 12GPT-6.1 Sol76.2
  14. 15GPT-5.6 Terra75.0
  15. 18GPT-6.1 Terra73.8
  16. 18GPT-6 Sol73.8
  17. 21o372.5
  18. 21GPT-5.472.5
  19. 21GPT-6 Luna72.5
  20. 41gpt-oss-120b58.8
  21. 46gpt-oss-20b51.2

51 more models · see the full leaderboard

6 of 78 tested models have no verification score

Output speed

OpenAI models highlighted against every model in this run · higher is better

  1. 1Nemotron 3.5 Super VL715
  2. 2Gemini 3.5 Flash Lite409
  3. 3Trinity Large Thinking329
  4. 4Ministral 3 3B 2512252
  5. 5Qwen3.5 397B A17B240
  6. 6Inkling Small233
  7. 7Muse Spark Open220
  8. 8Claude Haiku 5.5192
  9. 9Nova Micro 1.0180
  10. 10Muse Wonder175
  11. 14o3163
  12. 19GPT-6 Luna137
  13. 28GPT-6.1 Terra122
  14. 36gpt-oss-20b103
  15. 38GPT-5.6 Terra101
  16. 40GPT-5.3-Codex95
  17. 42GPT-6 Sol92
  18. 46gpt-oss-120b82
  19. 58GPT-5.455
  20. 59GPT-6 Astra49
  21. 64GPT-6.1 Sol41

51 more models · see the full leaderboard

6 of 78 tested models have no output speed

Time to first token

OpenAI models highlighted against every model in this run · lower is better

  1. 1Ministral 3 3B 25120.3 s
  2. 2Ministral 3 8B 25120.3 s
  3. 3Hermes 4 405B0.3 s
  4. 4Kimi K2.7 Code0.4 s
  5. 5MiMo-V2.6-Flash0.4 s
  6. 6Qwen3.6 35B A3B0.4 s
  7. 7Qwen3 Coder Next0.4 s
  8. 8Mistral Medium 3.50.5 s
  9. 9Phi 40.5 s
  10. 10Nova Micro 1.00.5 s
  11. 20gpt-oss-120b0.5 s
  12. 35gpt-oss-20b0.7 s
  13. 44GPT-6 Sol0.9 s
  14. 49GPT-5.6 Terra1.1 s
  15. 52GPT-6 Luna1.2 s
  16. 53o31.2 s
  17. 55GPT-6 Astra1.3 s
  18. 57GPT-5.3-Codex1.4 s
  19. 60GPT-5.41.7 s
  20. 64GPT-6.1 Sol2.1 s
  21. 67GPT-6.1 Terra2.4 s
  22. 72GPT-6.1 Terra Pro8.5 s
  23. 74GPT-5.5 Pro33.8 s

51 more models · see the full leaderboard

4 of 78 tested models have no time to first token

Cost per question

OpenAI models highlighted against every model in this run · lower is better

  1. 1Granite 4.0 Micro$0.000051
  2. 2gpt-oss-20b$0.000066
  3. 3Granite 4.2 8B$0.000077
  4. 4Nova Micro 1.0$0.000087
  5. 5Mercury 2.5$0.00011
  6. 6Phi 4$0.00012
  7. 7gpt-oss-120b$0.00013
  8. 8GLM 5.3 Flash$0.00015
  9. 9Ministral 3 3B 2512$0.00019
  10. 10GPT-6 Luna$0.00025
  11. 31GPT-6.1 Terra$0.0013
  12. 37GPT-6 Sol$0.0024
  13. 40GPT-6.1 Sol$0.0029
  14. 50GPT-6.1 Terra Pro$0.005
  15. 55GPT-5.6 Terra$0.0063
  16. 57o3$0.0088
  17. 59GPT-5.3-Codex$0.01
  18. 63GPT-5.4$0.011
  19. 64GPT-6 Astra$0.013

53 more models · see the full leaderboard

6 of 78 tested models have no cost per question
  • LFM2.5-2.6B: no verification score (80 of 80 questions failed on the host's side)
  • Ling 3.0 Flash VL: no verification score (not yet run on the current question suite)
  • Qwen3.5-122B-A10B: no verification score (not yet run on the current question suite)
  • Qwen3 Next 80B A3B Thinking: no verification score (not yet run on the current question suite)
  • Nova Premier 1.0: no verification score (not yet run on the current question suite)
  • GPT-5.5 Pro: no verification score (not yet run on the current question suite)

End-to-end response time

OpenAI models highlighted against every model in this run · lower is better

  1. 1Nemotron 3.5 Super VL1.5 s
  2. 2Ministral 3 3B 25121.6 s
  3. 3Nova Micro 1.02.1 s
  4. 4Trinity Large Thinking2.7 s
  5. 5Gemini 3.5 Flash Lite2.8 s
  6. 6DeepSeek V4 Flash Vision Exp3.2 s
  7. 7Muse Spark Open3.2 s
  8. 8GLM 5.33.4 s
  9. 9Ministral 3 8B 25123.5 s
  10. 10Qwen3.5 397B A17B3.5 s
  11. 14o33.9 s
  12. 15gpt-oss-20b4.0 s
  13. 19GPT-6 Luna4.3 s
  14. 20GPT-6 Sol4.4 s
  15. 25GPT-5.3-Codex5.2 s
  16. 27GPT-5.6 Terra5.3 s
  17. 30gpt-oss-120b5.4 s
  18. 33GPT-6.1 Terra5.8 s
  19. 47GPT-5.48.7 s
  20. 51GPT-6 Astra8.9 s
  21. 53GPT-6.1 Terra Pro9.0 s
  22. 59GPT-6.1 Sol10.8 s
  23. 72GPT-5.5 Pro35.4 s

51 more models · see the full leaderboard

4 of 78 tested models have no end-to-end response time

All OpenAI results

Leaderboard
#
1GPT-5.3-Codex
OpenAI · served by OpenAI
82.566 / 80951.4 s$0.01$1.75 / $14400k
2GPT-6 Astra
OpenAI · served by OpenAI
76.261 / 80491.3 s$0.013$10 / $501.1M
3GPT-6.1 Terra Pro
OpenAI · served by OpenAI
76.261 / 80—8.5 s$0.005$1 / $51.1M
4GPT-6.1 Sol
OpenAI · served by OpenAI
76.261 / 80412.1 s$0.0029$2 / $101.1M
5GPT-5.6 Terra
OpenAI · served by OpenAI
75.060 / 801011.1 s$0.0063$2 / $121.1M
6GPT-6.1 Terra
OpenAI · served by OpenAI
73.859 / 801222.4 s$0.0013$1 / $51.1M
7GPT-6 Sol
OpenAI · served by OpenAI
73.859 / 80920.9 s$0.0024$2 / $101.1M
8o3
OpenAI · served by OpenAI
72.558 / 801631.2 s$0.0088$2 / $8200k
9GPT-5.4
OpenAI · served by OpenAI
72.558 / 80551.7 s$0.011$2.5 / $151.1M
10GPT-6 Luna
OpenAI · served by OpenAI
72.558 / 801371.2 s$0.00025$0.1 / $0.51.1M
11gpt-oss-120b
OpenAI · served by CoreWeave
58.847 / 80820.5 s$0.00013$0.15 / $0.6131k
12gpt-oss-20b
OpenAI · served by Darkbloom
51.241 / 801030.7 s$0.000066$0.018 / $0.09131k
13GPT-5.5 Pro
OpenAI · served by OpenAI
—0 / 0—33.8 s—$30 / $1801.1M

OpenAI: quick answers

What is the best OpenAI model?
GPT-5.3-Codex has the highest verification score of the OpenAI models the leaderboard tested: 82.5% (66 of 80 correct) in the 2026-10-27 run.
What is the fastest OpenAI model?
o3 is the fastest OpenAI model the leaderboard tested, at a median 163 output tokens per second.
What is the cheapest OpenAI model?
gpt-oss-20b has the lowest billed cost per question of the OpenAI models the leaderboard tested: $0.000066.
Which OpenAI models do you test?
GPT-5.3-Codex, GPT-6 Astra, GPT-6.1 Terra Pro, GPT-6.1 Sol, GPT-5.6 Terra, GPT-6.1 Terra, GPT-6 Sol, o3, GPT-5.4, GPT-6 Luna, gpt-oss-120b, gpt-oss-20b, GPT-5.5 Pro.