Lab vs Lab
Moonshot AI vs DeepSeek.
The club-level face-off: Club Scores, best models, squad depth, division titles and the price floor — every number computed from the model database.
| Moonshot | DeepSeek | Edge | |
|---|---|---|---|
| Club Score | 68.3 | 73.0 | DeepSeek |
| Best model | Kimi K2.6 | DeepSeek V4.1 Flash | DeepSeek |
| Top-3 average | 66.2 | 69.3 | DeepSeek |
| Squad size | 3 | 2 | Moonshot |
| Cheapest capable ($/1M blended) | 1.7 | 0.5 | DeepSeek |
| Open-weight share % | 100 | 100 | Draw |
| Division titles | 0 | 1 | DeepSeek |
DeepSeek holds the #1 model in: Open-Weight Division.
Price floor — cheapest capable model on three workloads
| Workload | Moonshot | DeepSeek |
|---|---|---|
| Chat assistant (1M in / 250K out) | $1.95 | $0.60 |
| Document analysis (5M in / 500K out) | $6.75 | $2.10 |
| Code generation (2M in / 1000K out) | $5.90 | $1.80 |
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 | Kimi K2.6 | DeepSeek V4.1 Flash | Winner |
|---|---|---|---|
| League Score | 69.7 | 75.4 | DeepSeek V4.1 Flash |
| Coding (SWE-bench %) | 80.2 | — | — |
| Reasoning (GPQA %) | 90.5 | 90.9 | DeepSeek V4.1 Flash |
| Context (tokens) | 262144 | 1000000 | DeepSeek V4.1 Flash |
| Speed (tok/s) | — | 213 | — |
| Price (blended $/1M) | $1.71 | $0.52 | DeepSeek V4.1 Flash (cheaper) |
Tale of the tape
| Spec | Kimi K2.6 | DeepSeek V4.1 Flash |
|---|---|---|
| Maker | Moonshot AI | DeepSeek |
| Released | Apr 20, 2026 | Sep 10, 2026 |
| League Score | 70 | 75 |
| Context | 262,144 tokens | 1,000,000 tokens |
| Max output | — | 384,000 |
| $/1M input | $0.95 | $0.30 |
| $/1M output | $4.00 | $1.20 |
| Weights | Open | Open |
| Speed | — | 213 t/s |
Head-to-head benchmarks (shared only)
| Benchmark | Kimi K2.6 | DeepSeek V4.1 Flash |
|---|---|---|
| GPQA Diamond | 90.5 | 90.9 |
Cost at three workloads
| Workload | Kimi K2.6 | DeepSeek V4.1 Flash |
|---|---|---|
| Light — 1M in / 250K out | $1.95 | $0.60 |
| Standard — 10M in / 2.5M out | $19.50 | $6.00 |
| Long-context — 50M in / 5M out | $67.50 | $21.00 |
The pick
Pick Kimi K2.6 if coding is the priority.
Pick DeepSeek V4.1 Flash if you want the higher overall League Score; you need the bigger context window; 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.