GPT-5.4 minivsGemini 3.0 Flash
Across 6 shared benchmarks, Gemini 3.0 Flash leads overall: GPT-5.4 mini wins 2, Gemini 3.0 Flash wins 4, with 0 ties and an average score difference of +0.05.
GPT-5.4 mini
OpenAI · 2026-03-17 · Reasoning model
Gemini 3.0 Flash
Google Deep Mind · 2025-12-17 · Chat model
GPT-5.4 mini2 wins(33%)(67%)4 winsGemini 3.0 Flash
Benchmark scores
Grouped by capability, sorted by largest gap within each. 6 shared benchmarks.
General Knowledge
Gemini 3.0 Flash 2/2| Benchmark | GPT-5.4 mini | Gemini 3.0 Flash | Diff |
|---|---|---|---|
| GPQA Diamond | 8832 / 178极高强度思考（无工具） | 90.4017 / 178 | -2.40 |
| HLE | 41.5046 / 157极高强度思考（工具） | 43.5038 / 157 | -2 |
AI Agent - Tool Usage
GPT-5.4 mini 1/1| Benchmark | GPT-5.4 mini | Gemini 3.0 Flash | Diff |
|---|---|---|---|
| Terminal Bench 2.0 | 6019 / 46极高强度思考（工具） | 47.6037 / 46 | +12.40 |
Claw-style Agent Evaluation
Gemini 3.0 Flash 1/1| Benchmark | GPT-5.4 mini | Gemini 3.0 Flash | Diff |
|---|---|---|---|
| Claw Bench | 75.3025 / 29Thinking (With Tools) | 85.7015 / 29Thinking (With Tools) | -10.40 |
Coding and Software Engineer
GPT-5.4 mini 1/1| Benchmark | GPT-5.4 mini | Gemini 3.0 Flash | Diff |
|---|---|---|---|
| SWE-Bench Pro - Public | 54.4021 / 43极高强度思考（工具） | 49.6032 / 43Thinking High (With Tools) | +4.80 |
Math and Reasoning
Gemini 3.0 Flash 1/1| Benchmark | GPT-5.4 mini | Gemini 3.0 Flash | Diff |
|---|---|---|---|
| FrontierMath - Tier 4 | 2.1056 / 80Thinking High (No Tools) | 4.2040 / 80Normal (No Tools) | -2.10 |
Specs
| Field | GPT-5.4 mini | Gemini 3.0 Flash |
|---|---|---|
| Publisher | OpenAI | Google Deep Mind |
| Release date | 2026-03-17 | 2025-12-17 |
| Model type | Reasoning model | Chat model |
| Architecture | Dense | Dense |
| Parameters | Not available | Not available |
| Context length | 400K | 2000K |
| Max output | 128K | 64K |
API pricing
Prices use DataLearner records when available; missing fields are not inferred.
| Item | GPT-5.4 mini | Gemini 3.0 Flash |
|---|---|---|
| Text input | $0.75 / 1M tokens | Not public |
| Text output | $4.5 / 1M tokens | Not public |
| Cache read | $4.5 / 1M tokens | Not public |
| Cache write | $0.075 / 1M tokens | Not public |
One or both models have incomplete public pricing.
Summary
- GPT-5.4 minileads in:AI Agent - Tool Usage (1/1), Coding and Software Engineer (1/1)
- Gemini 3.0 Flashleads in:General Knowledge (2/2), Claw-style Agent Evaluation (1/1), Math and Reasoning (1/1)
On average across the 6 shared benchmarks, GPT-5.4 mini scores 0.05 higher.
Largest single-benchmark gap: Terminal Bench 2.0 — GPT-5.4 mini 60 vs Gemini 3.0 Flash 47.60 (+12.40).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.