Updated March 26, 2026· Based on independent benchmark data
MiniMax-M2.7 leads in intelligence with a score of 49.6 vs 45.0. MiniMax-M2.7 is 2.0x cheaper at $0.30/1M tokens vs $0.60/1M.
| Metric | MiniMax-M2.7 | Qwen3.5 397B A17B (Reasoning) |
|---|---|---|
| Intelligence Score | 49.6 | 45.0 |
| Coding Score | 41.9 | 41.3 |
| Math Score | N/A | N/A |
| Speed (tok/s) | 45 tok/s | 53 tok/s |
| Latency (TTFT) | 1.78s | 1.46s |
| Input Price / 1M tokens | $0.30 | $0.60 |
| Output Price / 1M tokens | $1.20 | $3.60 |
| Context Window | N/A |
MiniMax-M2.7 outperforms Qwen3.5 397B A17B (Reasoning) on the intelligence index with a score of 49.6 compared to 45.0. Their coding capabilities are closely matched (41.9 vs 41.3).
Both models deliver similar output speeds: MiniMax-M2.7 at 45 tok/s and Qwen3.5 397B A17B (Reasoning) at 53 tok/s. Time to first token is 1.46s for Qwen3.5 397B A17B (Reasoning) vs 1.78s for MiniMax-M2.7, which affects perceived responsiveness in interactive applications.
MiniMax-M2.7 is more affordable at $0.30/1M input tokens ($1.20/1M output), while Qwen3.5 397B A17B (Reasoning) costs $0.60/1M input ($3.60/1M output). For a typical workload of 100 requests per day at 2,000 tokens each, MiniMax-M2.7 would cost approximately $1.80/month vs $3.60/month for Qwen3.5 397B A17B (Reasoning) in input costs alone.
Choose MiniMax-M2.7 when you need higher intelligence (49.6), lower cost.
MiniMax-M2.7 scores higher on coding benchmarks (41.9 vs 41.3), making it the better choice for programming tasks.
MiniMax-M2.7 is cheaper at $0.30/1M input tokens vs $0.60/1M for Qwen3.5 397B A17B (Reasoning).
Qwen3.5 397B A17B (Reasoning) is faster, producing output at 53 tok/s compared to MiniMax-M2.7's 45 tok/s.
No, MiniMax-M2.7 does not support image input. Neither model supports image input.
Data last synced: March 26, 2026
| N/A |
| Max Output Tokens | N/A | N/A |
| Input Modalities | Text | Text |
| Output Modalities | Text | Text |
| Free Tier | No | No |
It depends on your priorities. MiniMax-M2.7 scores higher on intelligence (49.6), but Qwen3.5 397B A17B (Reasoning) may be better for specific use cases like budget-conscious projects or speed-critical applications.