Claude Opus 5 vs Kimi K3.
Verdict by category, tale of the tape, shared benchmarks only, and cost at three workloads — all from the tracked data.
Verdict by category
| Category | Claude Opus 5 | Kimi K3 | Winner |
|---|---|---|---|
| League Score | 94.6 | 70.6 | Claude Opus 5 |
| Coding (SWE-bench %) | 96.0 | — | — |
| Reasoning (GPQA %) | 93.2 | 93.5 | Kimi K3 |
| Context (tokens) | 1000000 | 1048576 | Kimi K3 |
| Speed (tok/s) | — | 44.8 | — |
| Price (blended $/1M) | $10.00 | $6.00 | Kimi K3 (cheaper) |
Tale of the tape
| Spec | Claude Opus 5 | Kimi K3 |
|---|---|---|
| Maker | Anthropic | Moonshot AI |
| Released | Jul 24, 2026 | Jul 16, 2026 |
| League Score | 95 | 71 |
| Context | 1,000,000 tokens | 1,048,576 tokens |
| Max output | 128,000 | 131,072 |
| $/1M input | $5.00 | $3.00 |
| $/1M output | $25.00 | $15.00 |
| Weights | Closed | Open |
| Speed | — | 44.8 t/s |
Head-to-head benchmarks (shared only)
| Benchmark | Claude Opus 5 | Kimi K3 |
|---|---|---|
| LiveBench overall | 82.1 | 64.6 |
| GPQA Diamond | 93.2 | 93.5 |
| LMArena Elo | 1522 | 1500 |
Cost at three workloads
| Workload | Claude Opus 5 | Kimi K3 |
|---|---|---|
| Light — 1M in / 250K out | $11.25 | $6.75 |
| Standard — 10M in / 2.5M out | $112.50 | $67.50 |
| Long-context — 50M in / 5M out | $375.00 | $225.00 |
The pick
Pick Claude Opus 5 if you want the higher overall League Score; coding is the priority.
Pick Kimi K3 if you need the bigger context window; you want the lower running cost; you want downloadable weights / self-hosting.
All figures from the model database — sources and dates on each model page. Cost your own workload or compare a different pair.
Frequently asked questions
Which is better overall, Claude Opus 5 or Kimi K3?
Claude Opus 5 leads on overall League Score, 95 to 71 — a gap of 24 points on a composite of sourced benchmarks. That is overall form, not every category: check the per-category verdicts before choosing.
Which is cheaper, Claude Opus 5 or Kimi K3?
Kimi K3 is cheaper: $6.00 vs $10.00 blended per 1M tokens (3:1 input/output mix) — about 1.7x less per token.
Which has the bigger context window, Claude Opus 5 or Kimi K3?
Kimi K3, with 1M of context against 1M for Claude Opus 5.
Which is better at reasoning, Claude Opus 5 or Kimi K3?
Kimi K3 leads on GPQA Diamond, 93.5% vs 93.2% on graduate-level science questions.
Are Claude Opus 5 and Kimi K3 open source?
Claude Opus 5 is closed weights — API access only; Kimi K3 is open weights — it can be downloaded and self-hosted.