GPT-5.4 benchmarks, pricing and specs
GPT-5.4 is an OpenAI model released on March 5, 2026 that takes text and image input and produces text. It costs $2.50 per 1M input tokens and $15 per 1M output tokens and accepts up to 1,050,000 tokens of context. It scores 1475 Elo on Arena Text, #18 of 39 current models (Arena, measured October 8, 2026).
- Per 1M input tokens
- $2.50
- Per 1M output tokens
- $15
- Context window
- 1.05M
- Arena Text
- 1475 Elo · #18/39
Data verified October 12, 2026 · Dataset 2026-10-12
Key facts
- Provider
- OpenAI
- Model family
- GPT-5.4
- Release date
- March 5, 2026
- Status
- Current
- Access
- API only
- API model id
- Not reported
- Context window
- 1,050,000 tokens
- Max output
- 128,000 tokens
- Input
- text, image
- Output
- text
- Task families
- text, multimodal
- Declared capabilities
- reasoning
GPT-5.4 pricing
It is the #11 most expensive of 37 current text models priced per 1M input tokens.
Independent benchmark results
Scores come from independent evaluators that ran the model themselves, pinned to a dated snapshot. Rank is among current models in the same snapshot. OpenAI's own launch figures are listed under strengths with the publisher named, because they are not measured the same way.
| Benchmark | Score | Rank | Measured | Source |
|---|---|---|---|---|
| Arena Text | 1475 Elo | #18 of 39 | October 8, 2026 | Arena · 2026-10-12 |
| GPQA Diamond | 93.3% | #13 of 37 | October 12, 2026 | Epoch AI · 2026-10-12 |
| FrontierMath Tiers 1-3 (v2) | 78.6% | #15 of 37 | October 12, 2026 | Epoch AI · 2026-10-12 |
| SimpleQA Verified | 45.1% | #23 of 34 | October 12, 2026 | Epoch AI · 2026-10-12 |
| LiveBench Overall | 77.97 points | #3 of 5 | June 25, 2026 | LiveBench · 2026-06-25 |
Strengths, cautions and best uses
Strengths
- GPT-5.4 provides a 1,050,000-token context window and configurable reasoning for complex professional work
Cautions
- Prompts above the provider long-context threshold use higher token rates; reproduce cost and quality on the target workload
Best for
- Long-context coding and professional-work evaluations
Alternatives to GPT-5.4
The closest current textmodels by price. The cheapest model that scores as well or better is listed first; failing that, the cheapest one within 3% of the benchmark's scale.
Cheaper, same or higher score
DeepSeek-V4.1-Flash
- 88% cheaper per 1M input tokens, $0.30 vs $2.50
- Same Arena Text score, 1475
- From DeepSeek
Kimi K3
- 20% more per 1M input tokens, $3 vs $2.50
- 13 points higher on Arena Text, 1488 vs 1475
- From Moonshot Kimi
Claude Sonnet 5
- 20% cheaper per 1M input tokens, $2 vs $2.50
- 14 points lower on Arena Text, 1461 vs 1475
- From Anthropic
Claude Sonnet 5.5
- 20% cheaper per 1M input tokens, $2 vs $2.50
- 1 points higher on Arena Text, 1476 vs 1475
- From Anthropic
Compare GPT-5.4 with
Articles about GPT-5.4
GPT-5.4 FAQ
How much does GPT-5.4 cost?
GPT-5.4 costs $2.50 per 1M input tokens and $15 per 1M output tokens at the standard tier. It is the #11 most expensive of 37 current text models priced per 1M input tokens. Source: OpenAI, verified August 7, 2026.
What is the context window of GPT-5.4?
GPT-5.4 accepts up to 1,050,000 tokens of context. It can return up to 128,000 tokens in one response. That ranks #1 of 44 current text models by context length.
How does GPT-5.4 score on independent benchmarks?
GPT-5.4 scores 1475 Elo on Arena Text, #18 of 39 current models (Arena, measured October 8, 2026), 93.3% on GPQA Diamond, #13 of 37 current models (Epoch AI, measured October 12, 2026) and 78.6% on FrontierMath Tiers 1-3 (v2), #15 of 37 current models (Epoch AI, measured October 12, 2026). These are independent evaluations, not OpenAI's own launch figures.
Is GPT-5.4 open source or API only?
GPT-5.4 is available through OpenAI's API only; its weights are not published.
What are the best alternatives to GPT-5.4?
The closest current alternatives by price and measured performance are DeepSeek-V4.1-Flash (88% cheaper per 1M input tokens, $0.30 vs $2.50), Kimi K3 (20% more per 1M input tokens, $3 vs $2.50), Claude Sonnet 5 (20% cheaper per 1M input tokens, $2 vs $2.50) and Claude Sonnet 5.5 (20% cheaper per 1M input tokens, $2 vs $2.50).
Sources
- GPT-5.4 model · OpenAI · official · verified August 7, 2026
- LiveBench 2026-06-25 task results · LiveBench · independent · verified August 7, 2026 · Apache-2.0
- Epoch AI Benchmarking Hub data (benchmark_data.zip) · Epoch AI · independent · verified October 12, 2026 · CC-BY-4.0
- Arena leaderboard dataset (lmarena-ai/leaderboard-dataset) · Arena · independent · verified October 12, 2026 · CC-BY-4.0
Cite this page
Van Data Team. "GPT-5.4 Benchmarks, Pricing and Specs." AI Model Benchmark Dataset 2026-10-12. https://vandatateam.com/ai-model-benchmarks/gpt-5-4
Reuse the data: JSON · CSV · or explore every model in the AI model benchmark comparison.
Production evaluation
Deciding whether GPT-5.4 fits your workload?
Public benchmarks narrow the shortlist. We test the finalists on your own tasks, data and cost limits, then build the routing and review controls to run the winner in production.
- Evaluation cases and acceptance thresholds tied to real work
- Model routing, fallbacks and cost controls