$ diff gpt-image-2-5-flare minimax-m3
GPT-Image-2.5 Flare vs MiniMax M3
Values come from GPT-Image-2.5 Flare and MiniMax M3, where each is cited to its source. This page states no benchmark result and ranks neither model — it puts two published specifications next to each other. Where a lab has not published a figure, the row says so rather than guessing.
OpenAIGPT-Image-2.5 Flare
- context
- —
- weights
- closed
- $/M in
- $30
- $/M out
- $2
- context
- 1M
- weights
- open
- $/M in
- $0.3
- $/M out
- $1.2
What actually differs
- Input price
- MiniMax M3 at $0.3/M against $30/M — 100.0× cheaper to feed. Standard rates: one of these labs also quotes a lower cached-input tier, which applies only when a prefix is reused — see the full spec below.
- Output price
- MiniMax M3 at $1.2/M against $2/M — 1.7× cheaper to generate. Output dominates the bill on most agentic workloads, where the model writes far more than it reads.
- Weights
- MiniMax M3 publishes weights (Open-weight (HuggingFace)); GPT-Image-2.5 Flare is API-only. That decides self-hosting, air-gapped deployment and fine-tuning before any capability question does.
- Recency
- GPT-Image-2.5 Flare shipped 99 days after MiniMax M3 (2026-09-08 vs 2026-06-01).
Full spec
| Attribute | GPT-Image-2.5 Flare | MiniMax M3 |
|---|---|---|
| Developer | OpenAI | MiniMax |
| Released | 2026-09-08 | 2026-06-01 |
| Context window | not recorded | 1,000,000 tokens (1M); min. 512K guaranteed |
| Pricing | image input $8.00/M tokens ($2.00/M cached) · image output $30.00/M tokens · text input $5.00/M tokens ($1.25/M cached) | $0.30/M input · $1.20/M output ($0.06/M cached input) — OpenRouter catalogue, read 2026-07-28 |
| License | proprietary | Open-weight (HuggingFace) |
| Availability | OpenAI API | API (global, including English) + open-weight on HuggingFace |