MiniMax M3 vs Qwen3.5 397B-A17B
| Spec | MiniMax M3 | Qwen3.5 397B-A17B |
|---|---|---|
| Respan Index | 31.2 (#76) | 29.8 (#82) |
| Blended price per 1M | $0.394 | $0.387 |
| Context window | 1M | 262K |
| Input / output per 1M | $0.225 / $0.9 | $0.172 / $1.03 |
| Max output | - | - |
| Input | text, vision | text, vision |
| Released | 2026-06-01 | 2026-02-15 |
Respan Index by category
Each category is scored 0 to 100, and how far ahead the winner is. Leads under a point are close to even.
| Category | MiniMax M3 | Qwen3.5 397B-A17B |
|---|---|---|
| Coding | 9.4 MiniMax M3 | 11.4 Qwen3.5 397B-A17B |
| Agents | 47.3 MiniMax M3 | 34.2 Qwen3.5 397B-A17B |
| Reasoning | 23.0 MiniMax M3 | 31.3 Qwen3.5 397B-A17B |
| Knowledge & language | 33.4 MiniMax M3 | 29.6 Qwen3.5 397B-A17B |
| Human preference | 52.7 MiniMax M3 | 50.7 Qwen3.5 397B-A17B |
Benchmarks both report
Qwen3.5 397B-A17B leads on 4 of the 9 benchmarks both report. Higher is better. Each score links to where it was published.
| Benchmark | MiniMax M3 | Qwen3.5 397B-A17B |
|---|---|---|
| SAGE 1.1 (Vals AI) | 85.5% | 85.5% |
| Chess Puzzles (Epoch AI) | 13% | 13% |
| LMArena Elo | 1439 | 1441 |
| LMArena WebDev (LMArena) | 1483 | 1400 |
| LMArena Vision (LMArena) | 1233 | 1246 |
| APEX-Agents (Mercor) | 37.7% | 24.9% |
| GPQA Diamond | 90.9% | 88.4% |
| Mystery Game Puzzles (Epoch AI) | 8% | 18% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 71.1% | 88.9% |
FAQ
Is MiniMax M3 better than Qwen3.5 397B-A17B?
MiniMax M3 scores 31.2 on the Respan Index (#76), ahead of Qwen3.5 397B-A17B at 29.8 (#82). Qwen3.5 397B-A17B leads on 4 of the 9 benchmarks both report.
Which is cheaper, MiniMax M3 or Qwen3.5 397B-A17B?
MiniMax M3 lists at $0.225 input and $0.9 output per 1M tokens; Qwen3.5 397B-A17B at $0.172 and $1.03. Both cost about the same per token at list price.
Which has the larger context window?
MiniMax M3 accepts 1M tokens and Qwen3.5 397B-A17B 262K tokens.