1
claude-opus-4-7-thinking
- Vendor
- Anthropic
- Score
- 1571
- Context
- 1M
- Pricing (in/out per 1M)
- $5 / $25

Late-April 2026 composite ranking of AI LLMs, including score, pricing, and context window.
The top four are all Anthropic Claude Opus 4.x models, led by claude-opus-4-7-thinking at 1571. Z.ai and Moonshot break into the top six; OpenAI, Google, Alibaba, and DeepSeek form the second tier.
| Rank | Model | Vendor | Score | Context | Pricing (in/out per 1M) |
|---|---|---|---|---|---|
| 1 | claude-opus-4-7-thinking | Anthropic | 1571 | 1M | $5 / $25 |
| 2 | claude-opus-4-7 |
1565 |
| 1M |
| $5 / $25 |
| 3 | claude-opus-4-6-thinking | Anthropic | 1551 | 1M | $5 / $25 |
| 4 | claude-opus-4-6 | Anthropic | 1548 | 1M | $5 / $25 |
| 5 | glm-5.1 | Z.ai | 1534 | 202.8K | $1.05 / $3.50 |
| 6 | kimi-k2.6 | Moonshot | 1529 | 256K | $0.74 / $4.66 |
| 7 | claude-sonnet-4-6 | Anthropic | 1525 | 1M | $3 / $15 |
| 8 | muse-spark | Meta | 1510 | — | — |
| 9 | gpt-5.5-high (codex-harness) | OpenAI | 1500 | — | — |
| 10 | claude-opus-4-5-20251101-thinking-32k | Anthropic | 1491 | 200K | $5 / $25 |
| 11 | qwen3.6-plus | Alibaba | 1470 | 1M | $0.33 / $1.95 |
| 12 | claude-opus-4-5-20251101 | Anthropic | 1468 | 200K | $5 / $25 |
| 13 | gpt-5.4-high (codex-harness) | OpenAI | 1457 | 1.1M | $2.50 / $15 |
| 14 | gemini-3.1-pro-preview | 1456 | 1M | $2 / $12 |
| 15 | deepseek-v4-pro-thinking | DeepSeek | 1455 | 1M | $0.43 / $0.87 |
| 16 | gpt-5.5 (codex-harness) | OpenAI | 1450 | — | — |
| 17 | glm-4.7 | Z.ai | 1440 | 202.8K | $0.38 / $1.74 |
| 18 | gemini-3-pro | 1438 | 1M | $2 / $12 |
| 19 | gpt-5.4-medium (codex-harness) | OpenAI | 1437 | 1.1M | $2.50 / $15 |
| 20 | gemini-3-flash | 1437 | 1M | $0.50 / $3 |
Comments