Lab vs Lab
DeepSeek vs Alibaba Qwen (Tongyi Lab).
The club-level face-off: Club Scores, best models, squad depth, division titles and the price floor — every number computed from the model database.
| DeepSeek | Qwen | Edge | |
|---|---|---|---|
| Club Score | 73.0 | 61.5 | DeepSeek |
| Best model | DeepSeek V4.1 Flash | Qwen3.8-Max | DeepSeek |
| Top-3 average | 69.3 | 61.2 | DeepSeek |
| Squad size | 2 | 2 | Draw |
| Cheapest capable ($/1M blended) | 0.5 | 3.0 | DeepSeek |
| Open-weight share % | 100 | 100 | Draw |
| Division titles | 1 | 0 | DeepSeek |
DeepSeek holds the #1 model in: Open-Weight Division.
Price floor — cheapest capable model on three workloads
| Workload | DeepSeek | Qwen |
|---|---|---|
| Chat assistant (1M in / 250K out) | $0.60 | $3.50 |
| Document analysis (5M in / 500K out) | $2.10 | $13.00 |
| Code generation (2M in / 1000K out) | $1.80 | $10.00 |
Best vs best
League fixture: The two clubs' #1 models, head-to-head. In Season Zero, fixture results are decided by the data — the category winners above. When voted fixtures launch, results blend 60% benchmark / 40% crowd per the published rulebook.
Verdict by category
| Category | DeepSeek V4.1 Flash | Qwen3.8-Max | Winner |
|---|---|---|---|
| League Score | 75.4 | 61.7 | DeepSeek V4.1 Flash |
| Coding (SWE-bench %) | — | — | — |
| Reasoning (GPQA %) | 90.9 | 92.6 | Qwen3.8-Max |
| Context (tokens) | 1000000 | 1000000 | Draw |
| Speed (tok/s) | 213 | 37 | DeepSeek V4.1 Flash |
| Price (blended $/1M) | $0.52 | $3.00 | DeepSeek V4.1 Flash (cheaper) |
Tale of the tape
| Spec | DeepSeek V4.1 Flash | Qwen3.8-Max |
|---|---|---|
| Maker | DeepSeek | Alibaba Qwen |
| Released | Sep 10, 2026 | Aug 3, 2026 |
| League Score | 75 | 62 |
| Context | 1,000,000 tokens | 1,000,000 tokens |
| Max output | 384,000 | 131,072 |
| $/1M input | $0.30 | $2.00 |
| $/1M output | $1.20 | $6.00 |
| Weights | Open | Open |
| Speed | 213 t/s | 37 t/s |
Head-to-head benchmarks (shared only)
| Benchmark | DeepSeek V4.1 Flash | Qwen3.8-Max |
|---|---|---|
| LiveBench overall | 77.3 | 64.2 |
| GPQA Diamond | 90.9 | 92.6 |
Cost at three workloads
| Workload | DeepSeek V4.1 Flash | Qwen3.8-Max |
|---|---|---|
| Light — 1M in / 250K out | $0.60 | $3.50 |
| Standard — 10M in / 2.5M out | $6.00 | $35.00 |
| Long-context — 50M in / 5M out | $21.00 | $130.00 |
The pick
Pick DeepSeek V4.1 Flash if you want the higher overall League Score; you want the lower running cost.
All figures from the model database — sources and dates on each model page. Cost your own workload or compare a different pair.