DeepSeek V5 (Leaked Staging) vs Qwen 2.5 Max
A direct, empirical evaluation of DeepSeek V5 (Leaked Staging) (DeepSeek AI) against Qwen 2.5 Max (Alibaba Cloud).
DeepSeek V5 (Leaked Staging)
Developed by DeepSeek AI · 1.8 Trillion (Sparse MoE) parameters
Autonomous coding agents scoring 81.4% on SWE-bench Verified, zero-telemetry self-hosting, sub-cent 1M-token processing, and formal mathematical proofs.
Qwen 2.5 Max
Developed by Alibaba Cloud · Proprietary (~700B) parameters
Cross-lingual document extraction and structured structured output generation.
| Metric | DeepSeek V5 (Leaked Staging) | Qwen 2.5 Max |
|---|---|---|
| SWE-bench Verified | 81.4% | 51.4% |
| MMLU-Pro | 87.5% | 77.2% |
| MATH-500 | 96.8% | 89.4% |
| Context Window | 1,000,000 tokens | 128,000 tokens |
| Input Pricing (per 1M) | $0.12 | $1.60 |
| Output Pricing (per 1M) | $0.48 | $6.40 |