GPT-5.4 vs Gemini 3.1 Pro
Side-by-side comparison of API pricing, specs, benchmarks, and capabilities
| Specification |
by OpenAI
|
by Google
|
|---|---|---|
| Overview | ||
| Status | Current flagship | Current flagship |
| Released | Aug 15, 2025 | Feb 19, 2026 |
| Pricing per million tokens | ||
| Input | $2.50/Mtok | $2.00/Mtok |
| Output | $15.00/Mtok | $12.00/Mtok |
| Blended avg | $5.63/Mtok | $4.50/Mtok |
| Cached input | $0.250/Mtok | $0.200/Mtok |
| Specifications | ||
| Context window | 1M tokens | 2M tokens |
| Parameters | Proprietary | Proprietary |
| Speed (TPS) | — | — |
| Modalities | ||
| Input |
textimage
|
textimageaudiovideo
|
| Benchmarks sources: OpenAI model card, Artificial Analysis / Vellum LLM Leaderboard | ||
| Avg benchmark score | 78.7 | 74.8 |
| Perf / dollar | 14 | 16.6 |
| GPQA Diamond | 88 | 94.3 |
| Humanity's Last Exam | 40.2 | 44.4 |
| ARC-AGI 2 | 50 | 77.1 |
| HumanEval | 94.5 | 80.6 |
| SWE-Bench Verified | 75 | — |
| Terminal-Bench 2.1 | — | 70.3 |
| MCP Atlas | — | 69.2 |
| BrowseComp | — | 85.9 |
| OSWorld-Verified | — | 76.2 |
| AIME 2025 | 95.5 | — |
| MMMLU | 88 | — |
| BFCL | 85 | — |
| MATH 500 | 92 | — |
| Independent composite scores each on its own scale — not part of the average above | ||
| AA Intelligence Index | 53.1 | 47.7 |
| AA Agentic Index | 44.2 | 23 |
| AA-Omniscience Index (−100–100) | 5.8 | 31.9 |
| LMSYS Chatbot Arena | 1465.5 ELO | 1486.5 ELO |
| Providers | ||
| Available from |
OpenAI — $2.50/$15.00/Mtok
|
Google — $2.00/$12.00/Mtok
|
Cost at scale
1M tokens · 50/50 input/output| Volume | GPT-5.4 | Gemini 3.1 Pro | Savings |
|---|---|---|---|
| 1M tokens | $5.63 | $4.5 | $1.13 (20.1%) |
| 10M tokens | $56.25 | $45 | $11.25 (20%) |
| 100M tokens | $562.5 | $450 | $112.5 (20%) |
| 1000M tokens | $5625 | $4500 | $1125 (20%) |
When to pick which
distilled from the pricing and spec data above- raw capability matters more than price: 78.7 vs 74.8 average benchmark score
- cost dominates: $4.50/Mtok blended vs $5.63 — 20% less on the same 50/50 token mix
- you need the longer context: 2M tokens vs 1M (2×)
The cheaper option here — Gemini 3.1 Pro costs $4.50/Mtok blended on Google.
Summary
GPT-5.4 by OpenAI costs $2.50/Mtok input and $15.00/Mtok output, with a 1M-token context window. It supports text, image input.
Gemini 3.1 Pro by Google costs $2.00/Mtok input and $12.00/Mtok output, with a 2M-token context window. It supports text, image, audio, video input.
On a blended cost basis, Gemini 3.1 Pro is 20% cheaper than GPT-5.4.It also has a larger context window.
On benchmarks, GPT-5.4 scores higher (78.7 vs 74.8) on average. In terms of value, Gemini 3.1 Pro has better performance per dollar (16.6 vs 14).
Note: Pricing is per million tokens. Actual costs vary with usage patterns, prompt caching, and batch discounts. Always verify against official provider pricing pages.