Kimi K3 Benchmark Details
Kimi K3 currently shows benchmark results led by BrowseComp (1 / 53, score 91.20), GPQA Diamond (8 / 187, score 93.50), AA-LCR (1 / 15, score 74.70). This page also compares it with 4 competitor models and 4 predecessor or same-series models, including performance and pricing views when available.
Benchmark Results
Benchmark Results
General Knowledge
4 evaluationsAI Agent - Information Search
2 evaluationsAI Agent - Tool Usage
8 evaluationsCoding and Software Engineer
8 evaluationsAgent Level Benchmark
4 evaluationsProductivity Knowledge
9 evaluationsMultimodal Understanding
14 evaluationsCompetitor Comparison
Benchmark scores for Kimi K3 compared against top models in its class
9 benchmarks with comparable scores. Each model shows its best score; mode label is displayed below.
| Benchmark | Kimi K3Current | GLM-5.2 | MiniMax M3 | Claude Opus 4.8 | GPT-5.6 Sol |
|---|---|---|---|---|---|
GPQA Diamond 综合评估 | 93.50Thinking Level · High | 91.20Thinking Enabled | -- | 93.60Thinking Level · High | -- |
HLE 综合评估 | 56.00Thinking Level · High ｜ Tools | 54.70Thinking Enabled ｜ Tools | -- | 57.90Extended Thinking ｜ Tools | -- |
BrowseComp AI Agent - 信息收集 | 91.20Thinking Level · High ｜ Tools | -- | 83.50Thinking Enabled ｜ Tools | 84.30Thinking Level · High ｜ Tools | -- |
MCP-Atlas AI Agent - 工具使用 | 84.20Thinking Level · High ｜ Tools | -- | -- | 82.20Deep Thinking Mode ｜ Tools | -- |
OSWorld 2.0 AI Agent - 工具使用 | 58.30Thinking Level · High ｜ Tools | -- | -- | -- | 62.60Thinking Level · Extra High ｜ Tools |
OSWorld-Verified AI Agent - 工具使用 | 84.80Thinking Level · High ｜ Tools | -- | 70.00Thinking Enabled ｜ Tools | 83.40Extended Thinking ｜ Tools | -- |
TerminalBench 2.1 AI Agent - 工具使用 | 88.30Thinking Level · High ｜ Tools | 81.00Thinking Level · High ｜ Tools | 66.00Thinking Enabled ｜ Tools | 78.90Thinking Level · High ｜ Tools | 88.80Thinking Level · High |
DeepSWE 编程与软件工程 | 67.50Thinking Level · High ｜ Tools | 44.00Deep Thinking Mode ｜ Tools | -- | 59.00Deep Thinking Mode ｜ Tools | 72.70Thinking Level · Extra High ｜ Tools |
Agents' Last Exam Agent能力评测 | 28.30Thinking Level · High ｜ Tools | -- | -- | -- | 52.70Thinking Level · Extra High ｜ Tools |
Standard API Pricing: Kimi K3 vs. Peer Models
Shows standard text input and output pricing side by side for each model. If extended-context pricing exists, the chart keeps the base rate and explains the threshold below.
Source: DataLearnerAI. Standard text prices shown here use the default supplier.
These models use different currencies or billing units, so the page falls back to raw price values instead of a shared bar chart.
| Model | Supplier | Standard input | Standard output | Base price applies to |
|---|---|---|---|---|
Kimi K3 | Moonshot AI | ¥20 / 1M tokens | ¥100 / 1M tokens | — |
GLM-5.2 | 智谱AI | $1.4 / 1M tokens | $4.4 / 1M tokens | — |
MiniMax M3 | MiniMaxAI | ¥2.1 / 1M tokens | ¥8.4 / 1M tokens | — |
Claude Opus 4.8 | Anthropic | $5 / 1M tokens | $25 / 1M tokens | — |
GPT-5.6 Sol | OpenAI | $5 / 1M tokens | $30 / 1M tokens | — |
Version History
How each version of the Kimi K3 series stacks up on benchmark tests
8 benchmarks with comparable scores. Each model shows its best score; mode label is displayed below.· Click a row to view its trend chart.
| Benchmark | Kimi K3Current | Kimi K2.7 Code | Kimi K2.6 | Kimi K2.5 | Kimi K2 Thinking |
|---|---|---|---|---|---|
GPQA Diamond 综合评估 | 93.50Thinking Level · High | -- | 90.50Thinking Enabled | 87.60Thinking Enabled | 84.50Thinking Enabled |
HLE 综合评估 | 56.00Thinking Level · High ｜ Tools | -- | 54.00Thinking Enabled ｜ Tools | 50.20Thinking Enabled ｜ Tools | 51.00Thinking Enabled ｜ Tools |
BrowseComp AI Agent - 信息收集 | 91.20Thinking Level · High ｜ Tools | -- | 83.20Thinking Enabled ｜ Tools | 60.60Thinking Enabled ｜ Tools | 60.20Thinking Enabled ｜ Tools |
AA-LCR 长上下文能力 | 74.70Thinking Level · High | -- | -- | 65.00Thinking Enabled | -- |
MCP-Atlas AI Agent - 工具使用 | 84.20Thinking Level · High ｜ Tools | -- | -- | 64.40Standard Mode ｜ Tools | -- |
OSWorld-Verified AI Agent - 工具使用 | 84.80Thinking Level · High ｜ Tools | -- | 73.10Thinking Enabled ｜ Tools | -- | -- |
TerminalBench 2.1 AI Agent - 工具使用 | 88.30Thinking Level · High ｜ Tools | 67.04Thinking Enabled ｜ Tools | 53.56Thinking Enabled | -- | -- |
DeepSWE 编程与软件工程 | 67.50Thinking Level · High ｜ Tools | 31.00Standard Mode ｜ Tools | -- | -- | -- |
Single-Benchmark Version Trend
Viewing: GPQA Diamond · 综合评估
Standard API Pricing Across the Kimi K3 Series
Shows standard text input and output pricing side by side for each model. If extended-context pricing exists, the chart keeps the base rate and explains the threshold below.
Source: DataLearnerAI. Standard text prices shown here use the default supplier.
These models use different currencies or billing units, so the page falls back to raw price values instead of a shared bar chart.
| Model | Supplier | Standard input | Standard output | Base price applies to |
|---|---|---|---|---|
Kimi K3 | Moonshot AI | ¥20 / 1M tokens | ¥100 / 1M tokens | — |
Kimi K2.7 Code | Moonshot AI | $0.95 / 1M tokens | $4 / 1M tokens | — |
Kimi K2.6 | Facebook AI研究实验室 | $0.95 / 1M tokens | $4 / 1M tokens | — |
Kimi K2.5 | Moonshot AI | $0.6 / 1M tokens | $3 / 1M tokens | — |
Kimi K2 Thinking | Fireworks AI | $0.6 / 1M tokens | $2.5 / 1M tokens | — |