MiniMax-M3
- Next shutdown
- None setNo date on 2 platforms
- Arena Text
- 1,441#51 of 128 · Arena
- GPQA Diamond
- 90.9%#28 of 108 · Epoch AI
- List price · $ per 1M
- $0.30 in $1.20 outTogether AI · cached input $0.06
- Context window
- 1Mtokens
Dates on every platform
2 listings · a floor is the earliest possible date
| Platform | Model ID and status | Deprecated | Shuts down | Replacement |
|---|---|---|---|---|
| Together AITogether AI: Changelog (opens in a new tab) · checked Oct 14, 2026 | MiniMaxAI/MiniMax-M3Active | – | None set | – |
| Fireworks AIFireworks AI: Minimax M3 model page (opens in a new tab) · checked Oct 14, 2026 | accounts/fireworks/models/minimax-m3Active | – | None set | – |
What breaks when you move off MiniMax-M3
No breaking changes are recorded for MiniMax-M3.
Intelligence
Third-party results, credited below
| Benchmark | Publisher | Score | Rank | Published |
|---|---|---|---|---|
| Arena Text | Arena | 1,441 ± 2 | #51 of 128 | Oct 8, 2026 |
| GPQA Diamond | Epoch AI | 90.9% ± 2.0% | #28 of 108 | Aug 10, 2026 |
| Arena WebDev | Arena | 1,482 ± 3 | #34 of 69 | Oct 8, 2026 |
| Arena Agent | Arena | -0 ± 0 | #34 of 35 | Oct 12, 2026 |
| OTIS Mock AIME 2024-2025 | Epoch AI | 71.1% ± 6.8% | #70 of 105 | Aug 10, 2026 |
Arena (arena.ai), https://arena.ai/leaderboard
Epoch AI, ‘Capabilities & benchmarking’. Published online at epoch.ai. Retrieved from ‘https://epoch.ai/benchmarks’ [online resource].
List prices
$ per 1M tokens, with each price’s source
| Platform | Input | Cached | Output | In effect | Source |
|---|---|---|---|---|---|
Together AIMiniMaxAI/MiniMax-M3 | $0.30 | $0.06 | $1.20 | Jun 12, 2026Current | Together AI: Available models (serverless) (opens in a new tab) |
Fireworks AIaccounts/fireworks/models/minimax-m3 | $0.30 | $0.06 | $1.20 | Oct 14, 2026Current · Standard serverless mode. Batch inference is billed at half the Standard rate; Priority, Fast and US-only endpoints cost more. | Fireworks AI: Serverless Pricing (opens in a new tab) |
Change history
ShutdownQwen3.5 397B leaves Together serverlessTogether AI1 break
Together removed Qwen/Qwen3.5-397B-A17B from serverless inference and recommends MiniMax-M3.
- Platform
- Together AI
- Model
- Qwen3.5-397B-A17B
- Replacement
- MiniMax-M3
- Announced
- AvailabilityRequests to the removed serverless model ID fail. The model can still run on an on-demand dedicated endpoint, billed per minute.
Source: Together AI, Changelog (opens in a new tab); Together AI, Deprecations (opens in a new tab)