Price-band ranking
Best AI models $2-10 / 1M tokens
As of November 7, 2026, Anthropic: Claude Opus 5.5 (batch) is the top-quality model priced $2-10 / 1m tokens, scoring 59.1 on the Artificial Analysis Intelligence Index among 22 eligible models in this price tier.
- #1 Anthropic: Claude Opus 5.5 (batch) - Quality 59.1 · $8.00/1M
- #2 OpenAI: GPT-6.1 Sol (batch) - Quality 54.6 · $4.00/1M
- #3 Anthropic: Claude Sonnet 5.5 (batch) - Quality 54.5 · $4.00/1M
- #4 Qwen: Qwen3.8 Max - Quality 53.4 · $3.00/1M
- #5 OpenAI: GPT-6.1 Terra (batch) - Quality 51.3 · $2.00/1M
- #6 Google: Gemini 4 Argon (batch) - Quality 50.4 · $4.00/1M
- #7 MoonshotAI: Kimi K3.1 Pro - Quality 46.5 · $4.00/1M
- #8 OpenAI: GPT-6 Sol (batch) - Quality 44.2 · $4.00/1M
| # | Model | Quality | Price/1M |
|---|---|---|---|
| 01 | Anthropic: Claude Opus 5.5 (batch) | 59.1 | $8.00 |
| 02 | OpenAI: GPT-6.1 Sol (batch) | 54.6 | $4.00 |
| 03 | Anthropic: Claude Sonnet 5.5 (batch) | 54.5 | $4.00 |
| 04 | Qwen: Qwen3.8 Max | 53.4 | $3.00 |
| 05 | OpenAI: GPT-6.1 Terra (batch) | 51.3 | $2.00 |
| 06 | Google: Gemini 4 Argon (batch) | 50.4 | $4.00 |
| 07 | MoonshotAI: Kimi K3.1 Pro | 46.5 | $4.00 |
| 08 | OpenAI: GPT-6 Sol (batch) | 44.2 | $4.00 |
| 09 | Meta: Muse Wonder | 43.6 | $8.00 |
| 10 | Meta: Muse Spark 1.3 | 36.6 | $2.00 |
| 11 | SpaceXAI: Grok 4.7 (batch) | 35.5 | $3.00 |
| 12 | OpenAI: GPT-5.3-Codex | 32.5 | $4.81 |
| 13 | OpenAI: GPT-5.6 Terra (batch) | 32.3 | $4.50 |
| 14 | SpaceXAI: Grok 4.6 (batch) | 31 | $3.00 |
| 15 | OpenAI: GPT-5.2 (batch) | 30.4 | $4.81 |
| 16 | OpenAI: GPT-5.2-Codex | 28.5 | $4.81 |
| 17 | Qwen: Qwen3.6 Max Preview | 28.4 | $2.31 |
| 18 | Meta: Muse Spark Open | 24.9 | $2.00 |
| 19 | OpenAI: GPT-5.1-Codex | 23.7 | $3.44 |
| 20 | MoonshotAI: Kimi K3 (batch) | 17.4 | $4.14 |
| 21 | Google: Gemini 2.5 Pro | 16.1 | $3.44 |
| 22 | Anthropic: Claude Haiku 4.5 (batch) | 7.4 | $2.00 |
Source: OpenRouter (openrouter.ai/rankings), as of November 7, 2026.
Benchmark scores: Artificial Analysis (artificialanalysis.ai) via OpenRouter (openrouter.ai/rankings).
Methodology: www.smophy.ai/benchmark/methodology
Token counts originate from each provider's own tokenizer and are not directly comparable across providers.