Model comparison - Open weights vs closed
GPT-6 Luna vs MiniMax-M3
GPT-6 Luna is from OpenAI and MiniMax-M3 is from MiniMax. On list price, MiniMax-M3 charges 3x GPT-6 Luna's input price ($0.30 vs $0.10 per 1M tokens), and MiniMax-M3 charges 2.4x GPT-6 Luna's output price ($1.20 vs $0.50 per 1M tokens). GPT-6 Luna has a 1,050,000-token context window and MiniMax-M3 1,000,000 tokens; max output is 128,000 and 524,288 tokens per response. GPT-6 Luna was released on September 22, 2026 and MiniMax-M3 on June 1, 2026. MiniMax-M3 publishes open weights (MiniMax Community License (Hugging Face license tag: other / minimax-community)); GPT-6 Luna is closed-weights.
Last verified from each provider's pricing page. All sources
Side-by-side specs
| Spec | GPT-6 Luna | MiniMax-M3 |
|---|---|---|
| Provider | OpenAI | MiniMax |
| Input / 1M tokens | $0.10 | $0.30 |
| Output / 1M tokens | $0.50 | $1.20 |
| Cache read / 1M | $0.01 | $0.06 |
| Cache write / 1M | $0.125 | - |
| Batch input / 1M | $0.05 | - |
| Batch output / 1M | $0.25 | - |
| Long-context rates above | 272,000 tokens | 512,000 tokens |
| Context window | 1,050,000 | 1,000,000 |
| Max output | 128,000 | 524,288 |
| Release date | ||
| Weights | Closed | Open |
| License | - | MiniMax Community License (Hugging Face license tag: other / minimax-community) |
| API model id | gpt-6-luna | MiniMax-M3 |
| AI Gateway id | openai/gpt-6-luna | minimax/minimax-m3 |
| Last verified |
Prices in USD per 1M tokens at each provider's standard first-party rate. A dash means the provider does not bill that item separately or no primary page states it.
What each costs in practice
The same three workloads every model page uses, worked from each model's standard list rates. Swap in your own token counts with the math on each model's page.
| Workload | GPT-6 Luna | MiniMax-M3 |
|---|---|---|
| One agent session1M input tokens and 100K output tokens, no caching. | $0.15 | $0.42 |
| A chat app10,000 requests a day at 2K input and 500 output tokens each. | $4.50$135.00 per 30-day month | $12.00$360.00 per 30-day month |
| The same session at a 90% cache hit90% of the input is read from the prompt cache. The one-time cache write premium is not included. | $0.069 | $0.204 |
Show the math
GPT-6 Luna
- One agent sessionInput: 1M x $0.10 = $0.10Output: 0.1M x $0.50 = $0.05
- A chat appInput: 20M x $0.10 = $2.00Output: 5M x $0.50 = $2.50
- The same session at a 90% cache hitCached input: 0.9M x $0.01 = $0.009Uncached input: 0.1M x $0.10 = $0.01Output: 0.1M x $0.50 = $0.05
MiniMax-M3
- One agent sessionInput: 1M x $0.30 = $0.30Output: 0.1M x $1.20 = $0.12
- A chat appInput: 20M x $0.30 = $6.00Output: 5M x $1.20 = $6.00
- The same session at a 90% cache hitCached input: 0.9M x $0.06 = $0.054Uncached input: 0.1M x $0.30 = $0.03Output: 0.1M x $1.20 = $0.12
Benchmarks under the same setting
No board in the benchmark directory lists both GPT-6 Luna and MiniMax-M3 under the same setting, so there is no like-for-like score to show here.
Set them up in your tools
Step-by-step guides whose official docs list GPT-6 Luna or MiniMax-M3 by model id.
- Use MiniMax in AiderLists MiniMax-M3
- Use MiniMax in Claude CodeLists MiniMax-M3
- Use MiniMax in ClineLists MiniMax-M3
- Use MiniMax in Codex CLILists MiniMax-M3
- Use MiniMax in CursorLists MiniMax-M3
- Use MiniMax in GitHub CopilotLists MiniMax-M3
- Use MiniMax in GooseLists MiniMax-M3
- Use MiniMax in Kilo CodeLists MiniMax-M3
- Use MiniMax in ZedLists MiniMax-M3
- Use OpenAI in AiderLists GPT-6 Luna
- Use OpenAI in ClineLists GPT-6 Luna
- Use OpenAI in Codex CLILists GPT-6 Luna
- Use OpenAI in ContinueLists GPT-6 Luna
- Use OpenAI in GooseLists GPT-6 Luna
- Use OpenAI in OpenCodeLists GPT-6 Luna
- Use OpenAI in Roo CodeLists GPT-6 Luna
- Use OpenAI in ZedLists GPT-6 Luna
Model pages
FAQ
How do GPT-6 Luna and MiniMax-M3 compare on price?
On list price, MiniMax-M3 charges 3x GPT-6 Luna's input price ($0.30 vs $0.10 per 1M tokens), and MiniMax-M3 charges 2.4x GPT-6 Luna's output price ($1.20 vs $0.50 per 1M tokens). For one agent session of 1M input and 100K output tokens with no caching, that is $0.15 on GPT-6 Luna and $0.42 on MiniMax-M3. Prices are each provider's first-party list rates, verified October 17, 2026 and October 17, 2026.
What are the context windows of GPT-6 Luna and MiniMax-M3?
GPT-6 Luna has a 1,050,000-token context window and MiniMax-M3 1,000,000 tokens; max output is 128,000 and 524,288 tokens per response.
Are GPT-6 Luna and MiniMax-M3 open weights?
GPT-6 Luna is closed-weights and is called through OpenAI's API as gpt-6-luna. MiniMax-M3's weights are published at huggingface.co under the MiniMax Community License (Hugging Face license tag: other / minimax-community).
More comparisons
- Claude Haiku 5.5 vs GPT-6 LunaSimilar price, same tier
- Claude Haiku 5.5 vs MiniMax-M3Open weights vs closed
- DeepSeek V4.1 Flash vs GPT-6 LunaOpen weights vs closed
- GLM-5.3-Flash vs GPT-6 LunaSimilar price, same tier
- DeepSeek V4.1 Flash vs MiniMax-M3Similar price, same tier
- GPT-6 Luna vs Qwen3.8-FlashSimilar price, same tier
- GLM-5.3-Flash vs MiniMax-M3Similar price, same tier
- GPT-5.6 Luna vs GPT-6 LunaSame provider, successor
Sources
GPT-6 Luna verified from developers.openai.com; MiniMax-M3 verified from platform.minimax.io. This page is as current as the older of the two: .
- https://developers.openai.com/api/docs/pricing
- https://platform.minimax.io/docs/guides/pricing
- https://developers.openai.com/api/docs/models/gpt-6-luna
- https://developers.openai.com/api/docs/changelog
- https://platform.minimax.io/docs/pricing/overview
- https://platform.minimax.io/docs/guides/text-generation
- https://platform.minimax.io/docs/api-reference/text-anthropic-api
- https://platform.minimax.io/docs/api-reference/text-chat-anthropic
- https://platform.minimax.io/docs/release-notes/models
- https://huggingface.co/MiniMaxAI/MiniMax-M3
- https://huggingface.co/MiniMaxAI/MiniMax-M3/raw/main/LICENSE
Newsletter
Get model price changes as they ship
One email per week. New model releases, price cuts and what they mean for your bill.