DeepSeek V4 Pro vs Kimi K3
As of Oct 15, 2026, DeepSeek V4 Pro leads Kimi K3 in 0 of 6 shared categories.
DeepSeek V4 Pro
#28 Best recorded rank: #28 in Science, as of Oct 15, 2026.
Kimi K3
Yes. Yes on General LLM (open weights); Reasoning (open weights). Not the overall leader.
| Category | DeepSeek V4 Pro | Kimi K3 |
|---|---|---|
| General LLM | #55 · #55 in General LLM Margin 29.255319148936167 LMArena leaderboard dataset · Oct 13, 2026 | #13 · #13 in General LLM Margin 6.914893617021278 LMArena leaderboard dataset · Oct 13, 2026 |
| Reasoning | #57 · #57 in Reasoning Margin 30.851063829787236 LMArena leaderboard dataset · Oct 13, 2026 | #7 · #7 in Reasoning Margin 4.255319148936167 LMArena leaderboard dataset · Oct 13, 2026 |
| Math | #— · Not enough data Margin — LMArena leaderboard dataset · Oct 13, 2026 | #— · Not enough data Margin — LMArena leaderboard dataset · Oct 13, 2026 |
| Science | #28 · #28 in Science Margin — Epoch AI benchmarking hub · Apr 24, 2026 | #17 · #17 in Science Margin — Epoch AI benchmarking hub · Jul 16, 2026 |
| Coding | #— · Not enough data Margin — LMArena leaderboard dataset · Oct 13, 2026 | #— · Not enough data Margin — Scale Labs leaderboards · Oct 14, 2026 |
| Coding › Web dev | #62 · #62 in Coding › Web dev Margin 48.8 LMArena leaderboard dataset · Oct 13, 2026 | #12 · Was SOTA · Coding › Web dev Margin 8.799999999999997 LMArena leaderboard dataset · Oct 13, 2026 |
| Long context | #51 · #51 in Long context Margin 27.127659574468083 LMArena leaderboard dataset · Oct 13, 2026 | #8 · Was SOTA · Long context Margin 4.255319148936167 LMArena leaderboard dataset · Oct 13, 2026 |
| Vision | — | #22 · #22 in Vision Margin 60.526315789473685 Epoch AI benchmarking hub · Jul 16, 2026 |
| Factuality | #49 · #49 in Factuality Margin 29.069767441860463 LMArena leaderboard dataset · Oct 13, 2026 | #22 · #22 in Factuality Margin 13.372093023255815 LMArena leaderboard dataset · Oct 13, 2026 |
| Abstraction (ARC) | — | #8 · #8 in Abstraction (ARC) Margin — Epoch AI benchmarking hub · Jul 16, 2026 |
DeepSeek V4 Pro: $1.194075 per 1M blended tokens.
Kimi K3: $4.109999999999999 per 1M blended tokens.
DeepSeek V4 Pro: open weights.
Kimi K3: open weights.
General LLM: As of Oct 15, 2026, Kimi K3 leads DeepSeek V4 Pro in General LLM at #13 against #55.
Reasoning: As of Oct 15, 2026, Kimi K3 leads DeepSeek V4 Pro in Reasoning at #7 against #57.
Science: As of Oct 15, 2026, Kimi K3 leads DeepSeek V4 Pro in Science at #17 against #28.
Coding › Web dev: As of Oct 15, 2026, Kimi K3 leads DeepSeek V4 Pro in Coding › Web dev at #12 against #62.
Long context: As of Oct 15, 2026, Kimi K3 leads DeepSeek V4 Pro in Long context at #8 against #51.
Factuality: As of Oct 15, 2026, Kimi K3 leads DeepSeek V4 Pro in Factuality at #22 against #49.
DeepSeek V4 Pro or Kimi K3?
As of Oct 15, 2026, DeepSeek V4 Pro leads Kimi K3 in 0 of 6 shared categories.
DeepSeek V4 Pro or Kimi K3 for General LLM?
As of Oct 15, 2026, Kimi K3 leads DeepSeek V4 Pro in General LLM at #13 against #55.
DeepSeek V4 Pro or Kimi K3 for Reasoning?
As of Oct 15, 2026, Kimi K3 leads DeepSeek V4 Pro in Reasoning at #7 against #57.
DeepSeek V4 Pro or Kimi K3 for Science?
As of Oct 15, 2026, Kimi K3 leads DeepSeek V4 Pro in Science at #17 against #28.