OpenAI: Models Intelligence, Performance & Price
Analysis of OpenAI's models across key metrics including quality, price, output speed, latency, context window & more. This analysis is intended to support you in choosing the best model provided by OpenAI for your use-case.
Most Intelligent
Intelligence index
Total 61 models
Fastest
Output speed
Total 61 models
Lowest Price
Blended price (per 1M tokens)
Total 61 models
OpenAI offers 61 models, each with different intelligence, performance, and pricing characteristics. Below is a comparison of the key metrics across models.
- For intelligence, the top models on OpenAI are GPT-5.6 Sol (max) (61), GPT-5.6 Sol (xhigh) (59), and GPT-5.6 Sol (high) (57).
- For output speed, the fastest models are o3-mini (high) (228 t/s), o3-mini (215 t/s), and GPT-5.6 Luna (medium) (187 t/s).
- For latency, GPT-4.1 nano (0.67s), GPT-5.4 nano (0.73s), and GPT-5.4 mini (0.74s) offer the lowest time to first answer token.
- For pricing, GPT-5 nano (high) ($0.05), GPT-5 nano (medium) ($0.05), and GPT-5 nano (minimal) ($0.05) offer the lowest blended prices per 1M tokens. Prices vary up to 2.7x across models.
- For context window size, GPT-5.5 (xhigh) (1M), GPT-5.5 (high) (1M), and GPT-5.4 (xhigh) (1M) support the largest context windows on OpenAI.
Highlights
Intelligence Evaluations
Artificial Analysis Intelligence Index
Intelligence Evaluations
Agentic real-world work tasks, (Elo-500)/2000
Agentic tool use
Agentic coding & terminal use
Coding
Reasoning & knowledge
Scientific reasoning
Physics reasoning
Knowledge
1 - hallucination rate
Long context reasoning
Agentic knowledge work, Elo
Agentic SaaS workflows
Legal agentic work, criterion pass rate
Agentic business operations
Instruction following
Long-horizon agentic tasks
Kubernetes incident root-cause analysis
Visual reasoning
Intelligence Index vs. Price
Context Window
Context Window
Pricing
Intelligence Index vs. Price
Performance Summary
Output Speed vs. Price
Speed
Measured by Output Speed (tokens per second)
Output Speed
Latency
Measured by Time (seconds) to First Token
Latency: Time To First Answer Token
End-to-End Response Time
Seconds to output 500 tokens, calculated based on time to first token, 'thinking' time for reasoning models, and output speed
End-to-End Response Time vs. Price
Further Analysis | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
GPT-5.6 Sol (max) | 1M | Proprietary | 61 | $1.23 | 63 | 120.03 | 127.93 | -- | |||
GPT-5.6 Sol (xhigh) | 1M | Proprietary | 59 | $0.81 | 64 | 55.51 | 63.36 | -- | |||
GPT-5.6 Sol (high) | 1M | Proprietary | 57 | $0.55 | 61 | 13.86 | 22.02 | -- | |||
GPT-5.6 Terra (max) | 1M | Proprietary | 57 | $0.51 | 132 | 143.79 | 147.59 | -- | |||
GPT-5.5 (xhigh) | 1.05M | Proprietary | 56 | $1.17 | 72 | 61.30 | 68.20 | -- | |||
GPT-5.6 Sol (medium) | 1M | Proprietary | 56 | $0.37 | 60 | 4.84 | 13.14 | -- | |||
GPT-5.5 (high) | 1.05M | Proprietary | 55 | $0.80 | 68 | 20.09 | 27.40 | -- | |||
GPT-5.4 (xhigh) | 1.05M | Proprietary | 53 | $1.10 | 124 | 102.64 | 106.67 | -- | |||
GPT-5.6 Terra (xhigh) | 1M | Proprietary | 53 | $0.31 | 120 | 11.98 | 16.15 | -- | |||
GPT-5.6 Luna (max) | 1M | Proprietary | 52 | $0.05 | 176 | 83.59 | 86.44 | -- | |||
GPT-5.5 (medium) | 1.05M | Proprietary | 51 | $0.50 | 69 | 7.28 | 14.49 | -- | |||
GPT-5.6 Sol (low) | 1M | Proprietary | 51 | $0.23 | 62 | 3.45 | 11.51 | -- | |||
GPT-5.6 Terra (high) | 1M | Proprietary | 50 | $0.22 | 115 | 2.44 | 6.79 | -- | |||
GPT-5.6 Luna (xhigh) | 1M | Proprietary | 50 | $0.03 | 178 | 31.03 | 33.84 | -- | |||
GPT-5.6 Luna (high) | 1M | Proprietary | 47 | $0.02 | 181 | 8.13 | 10.89 | -- | |||
GPT-5.6 Terra (medium) | 1M | Proprietary | 47 | $0.12 | 113 | 1.60 | 6.01 | -- | |||
GPT-5.3 Codex (xhigh) | 400k | Proprietary | 46* | -- | 130 | 74.01 | 77.86 | -- | |||
GPT-5.5 (low) | 1.05M | Proprietary | 44 | $0.26 | 72 | 1.72 | 8.70 | -- | |||
GPT-5.2 (xhigh) | 400k | Proprietary | 43* | -- | 77 | 132.23 | 138.73 | -- | |||
GPT-5.6 Sol (Non-reasoning) | 1M | Proprietary | 42 | $0.24 | 58 | 1.06 | 9.74 | -- | |||
GPT-5.6 Terra (low) | 1M | Proprietary | 41 | $0.09 | 121 | 1.51 | 5.65 | -- | |||
GPT-5.4 mini (xhigh) | 400k | Proprietary | 41 | $0.49 | 178 | 4.92 | 7.73 | -- | |||
GPT-5.4 (low) | 1.05M | Proprietary | 40* | -- | 99 | 1.93 | 6.97 | -- | |||
GPT-5.4 nano (xhigh) | 400k | Proprietary | 40 | $0.15 | 154 | 4.86 | 8.11 | -- | |||
GPT-5.2 (medium) | 400k | Proprietary | 39* | -- | 77 | 6.40 | 12.90 | -- | |||
GPT-5.6 Luna (medium) | 1M | Proprietary | 39 | $0.01 | 187 | 2.22 | 4.89 | -- | |||
GPT-5.1 (high) | 272k | Proprietary | 37 | $0.30 | 92 | 45.58 | 51.03 | -- | |||
GPT-5.5 (Non-reasoning) | 1.05M | Proprietary | 36 | $0.21 | 65 | 1.05 | 8.73 | -- | |||
GPT-5 (high) | 400k | Proprietary | 35 | $0.26 | 98 | 85.10 | 90.19 | -- | |||
GPT-5.6 Terra (Non-reasoning) | 1M | Proprietary | 35 | $0.10 | 111 | 0.77 | 5.28 | -- | |||
GPT-5 (medium) | 400k | Proprietary | 35* | -- | 78 | 34.67 | 41.11 | -- | |||
GPT-5.5 Instant (May 2026) | 400k | Proprietary | 34* | -- | -- | -- | -- | -- | |||
GPT-5.6 Luna (low) | 1M | Proprietary | 34 | $0.01 | 168 | 1.94 | 4.91 | -- | |||
o3-pro | 200k | Proprietary | 33* | -- | -- | -- | -- | -- | |||
GPT-5 (low) | 400k | Proprietary | 32* | -- | 82 | 8.27 | 14.33 | -- | |||
GPT-5 mini (medium) | 400k | Proprietary | 32* | -- | 83 | 15.98 | 22.04 | -- | |||
o3 | 200k | Proprietary | 31* | -- | -- | -- | -- | -- | |||
o3 | 200k | Proprietary | 31* | -- | 127 | 7.98 | 11.91 | -- | |||
GPT-5.4 nano | 400k | Proprietary | 31* | -- | 153 | 5.86 | 9.12 | -- | |||
GPT-5.4 mini (medium) | 272k | Proprietary | 30* | -- | 170 | 6.00 | 8.94 | -- | |||
GPT-5.5 Instant (June 2026) | 400k | Proprietary | 29 | $0.54 | 161 | 1.13 | 16.64 | 12.41 | |||
GPT-5.4 (Non-reasoning) | 1.05M | Proprietary | 28* | -- | 94 | 0.94 | 6.26 | -- | |||
GPT-5.6 Luna (Non-reasoning) | 1M | Proprietary | 27 | $0.01 | 160 | 0.84 | 3.96 | -- | |||
GPT-5.2 | 400k | Proprietary | 27* | -- | 71 | 1.16 | 8.23 | -- | |||
o4-mini (high) | 200k | Proprietary | 26* | -- | -- | -- | -- | -- | |||
o4-mini (high) | 200k | Proprietary | 26* | -- | 149 | 26.50 | 29.86 | -- | |||
GPT-5 mini (high) | 400k | Proprietary | 26 | $0.04 | 89 | 74.53 | 80.16 | -- | |||
o1 | 200k | Proprietary | 24* | -- | -- | -- | -- | -- | |||
GPT-5.1 | 400k | Proprietary | 21* | -- | 92 | 1.04 | 6.50 | -- | |||
GPT-5 nano (high) | 400k | Proprietary | 20* | -- | 150 | 104.70 | 108.03 | -- | |||
GPT-4.1 | 1M | Proprietary | 20* | -- | 105 | 1.01 | 5.77 | -- | |||
GPT-5 nano (medium) | 400k | Proprietary | 19* | -- | 152 | 46.54 | 49.82 | -- | |||
o3-mini | 200k | Proprietary | 19* | -- | 215 | 5.68 | 8.01 | -- | |||
o1-pro | 200k | Proprietary | 19* | -- | -- | -- | -- | -- | |||
GPT-5.4 nano | 400k | Proprietary | 18* | -- | 153 | 0.73 | 4.00 | -- | |||
GPT-5 (minimal) | 400k | Proprietary | 17* | -- | 69 | 1.31 | 8.52 | -- | |||
GPT-5 (minimal) private | 400k | Proprietary | 17* | -- | -- | -- | -- | -- | |||
GPT-5.4 mini | 272k | Proprietary | 17* | -- | 155 | 0.74 | 3.96 | -- | |||
o3-mini (high) | 200k | Proprietary | 16 | -- | 228 | 19.78 | 21.97 | -- | |||
GPT-4.1 mini | 1M | Proprietary | 15 | $0.06 | 70 | 0.94 | 8.06 | -- | |||
GPT-5 mini (minimal) | 400k | Proprietary | 14* | -- | 87 | 1.12 | 6.86 | -- | |||
GPT-5 mini (minimal) private | 400k | Proprietary | 14* | -- | -- | -- | -- | -- | |||
GPT-4o (Nov) | 128k | Proprietary | 11* | -- | 125 | 0.86 | 4.86 | -- | |||
GPT-4.1 nano | 1M | Proprietary | 10 | $0.03 | 152 | 0.67 | 3.96 | -- | |||
GPT-4o (Aug) | 128k | Proprietary | 9* | -- | 83 | 1.07 | 7.05 | -- | |||
GPT-4o (Aug) | 128k | Proprietary | 9* | -- | -- | -- | -- | -- | |||
GPT-4o (May) | 128k | Proprietary | 8* | -- | 70 | 1.38 | 8.48 | -- | |||
GPT-5 nano (minimal) | 400k | Proprietary | 8* | -- | 144 | 0.85 | 4.33 | -- | |||
GPT-5 nano (minimal) private | 400k | Proprietary | 8* | -- | -- | -- | -- | -- | |||
GPT-4 Turbo | 128k | Proprietary | 8* | -- | 31 | 3.02 | 19.11 | -- | |||
GPT-4 | 8.19k | Proprietary | 7* | -- | -- | -- | -- | -- | |||
GPT-4o mini | 128k | Proprietary | 7* | -- | -- | -- | -- | -- | |||
GPT-4o mini | 128k | Proprietary | 7* | -- | 86 | 1.25 | 7.05 | -- | |||
GPT-3.5 Turbo | 4.1k | Proprietary | 3* | -- | -- | -- | -- | -- | |||
GPT-5.4 Pro (xhigh) | 1.05M | Proprietary | -- | -- | -- | -- | -- | -- | |||
Key definitions
Frequently Asked Questions
Common questions about OpenAI
OpenAI offers 61 models that we track: GPT-5.6 Sol (max), GPT-5.6 Sol (xhigh), GPT-5.6 Sol (high), GPT-5.6 Terra (max), GPT-5.5 (xhigh), GPT-5.6 Sol (medium), GPT-5.5 (high), GPT-5.4 (xhigh), GPT-5.6 Terra (xhigh), GPT-5.6 Luna (max), GPT-5.5 (medium), GPT-5.6 Sol (low), GPT-5.6 Terra (high), GPT-5.6 Luna (xhigh), GPT-5.6 Luna (high), GPT-5.6 Terra (medium), GPT-5.3 Codex (xhigh), GPT-5.5 (low), GPT-5.2 (xhigh), GPT-5.6 Sol (Non-reasoning), GPT-5.6 Terra (low), GPT-5.4 mini (xhigh), GPT-5.4 (low), GPT-5.4 nano (xhigh), GPT-5.2 (medium), GPT-5.6 Luna (medium), GPT-5.1 (high), GPT-5.5 (Non-reasoning), GPT-5 (high), GPT-5.6 Terra (Non-reasoning), GPT-5 (medium), GPT-5.6 Luna (low), GPT-5 (low), GPT-5 mini (medium), o3, GPT-5.4 nano, GPT-5.4 mini (medium), GPT-5.5 Instant (June 2026), GPT-5.4 (Non-reasoning), GPT-5.6 Luna (Non-reasoning), GPT-5.2, o4-mini (high), GPT-5 mini (high), GPT-5.1, GPT-5 nano (high), GPT-4.1, GPT-5 nano (medium), o3-mini, GPT-5.4 nano, GPT-5 (minimal), GPT-5.4 mini, o3-mini (high), GPT-4.1 mini, GPT-5 mini (minimal), GPT-4o (Nov), GPT-4.1 nano, GPT-4o (Aug), GPT-4o (May), GPT-5 nano (minimal), GPT-4 Turbo, and GPT-4o mini.
The most intelligent model available on OpenAI is GPT-5.6 Sol (max) with an Intelligence Index score of 61.
The fastest model on OpenAI by output speed is o3-mini (high) at 228.0 tokens per second.
The model with the lowest time to first answer token on OpenAI is GPT-4.1 nano at 0.67s. Lower latency means faster initial response time.
The most affordable model on OpenAI by blended price is GPT-5 nano (high) at $0.05 per 1M tokens (7:2:1 cache hit/input/output ratio).
Prices on OpenAI vary up to 224x across models, from $0.05 per 1M tokens for GPT-5 nano (high) to $12.00 per 1M tokens for GPT-4 Turbo.
Yes, OpenAI offers an OpenAI-compatible API, making it easy to switch from OpenAI or use existing OpenAI SDK integrations.
Yes, all 61 models on OpenAI support JSON mode for structured output.
Yes, all 61 models on OpenAI support function calling (tool use).
Yes, OpenAI offers 41 reasoning models: GPT-5.6 Sol (max), GPT-5.6 Sol (xhigh), GPT-5.6 Sol (high), GPT-5.6 Terra (max), GPT-5.5 (xhigh), GPT-5.6 Sol (medium), GPT-5.5 (high), GPT-5.4 (xhigh), GPT-5.6 Terra (xhigh), GPT-5.6 Luna (max), GPT-5.5 (medium), GPT-5.6 Sol (low), GPT-5.6 Terra (high), GPT-5.6 Luna (xhigh), GPT-5.6 Luna (high), GPT-5.6 Terra (medium), GPT-5.3 Codex (xhigh), GPT-5.5 (low), GPT-5.2 (xhigh), GPT-5.6 Terra (low), GPT-5.4 mini (xhigh), GPT-5.4 (low), GPT-5.4 nano (xhigh), GPT-5.2 (medium), GPT-5.6 Luna (medium), GPT-5.1 (high), GPT-5 (high), GPT-5 (medium), GPT-5.6 Luna (low), GPT-5 (low), GPT-5 mini (medium), o3, GPT-5.4 nano, GPT-5.4 mini (medium), GPT-5.5 Instant (June 2026), o4-mini (high), GPT-5 mini (high), GPT-5 nano (high), GPT-5 nano (medium), o3-mini, and o3-mini (high). Reasoning models use extended thinking to work through complex problems before providing an answer.
Yes, provider performance can vary over time due to infrastructure changes, load balancing, and updates. We continuously benchmark all providers and display historical performance trends in the "Over Time" charts.
When choosing a model on OpenAI, consider: intelligence (for quality-sensitive tasks), output speed (for throughput-intensive tasks), latency (for interactive applications requiring quick first responses), pricing (for cost-sensitive workloads), and features like context window size, JSON mode, or function calling support.