Reported by others
Professional work tasks (APEX-Agents)
Investment banking, consulting and legal work that the model finishes in a simulated workplace with files and apps.
How to read this table
- Higher rows are rated higher. A range like “1–2” means the gap is within the margin of error.
- The score is the value published by the task's authors and other organizations.
- Conditions can differ between models, so small gaps do not mean much.
- Prices are a guide per million tokens, reading then writing.
- Length is how much text the model can read at once (tokens).
± range of the top model
| Rank | Model | Developer | Success rate (%) | Input $ /1M | Output $ /1M | Context | Supports | Weights | Released | Price and specs |
|---|---|---|---|---|---|---|---|---|---|---|
| 1–3 | Gemini 4 Argon | Rating 82.4 ±8.4 | $2 | $10 | 1.05M | Google Released 2026-09 Price $2 / $10 Length 1.05M | ||||
| 1–11 | Claude Sonnet 5.5 | Anthropic | Rating 75.5 ±9.2 | $2 | $10 | 1M | Anthropic Released 2026-09 Price $2 / $10 Length 1M | |||
| 1–12 | Claude Opus 5.5 | Anthropic | Rating 73.5 ±9.6 | $4 | $20 | 1M | Anthropic Released 2026-09 Price $4 / $20 Length 1M | |||
| 2–20 | Claude Fable 5.1 | Anthropic | Rating 68.6 ±9.6 | $10 | $50 | 1M | Anthropic Released 2026-09 Price $10 / $50 Length 1M | |||
| 2–22 | Gemini 3.7 Flash | Rating 67.8 ±10.0 | $0.75 | $3.75 | 1.05M | Google Released 2026-08 Price $0.75 / $3.75 Length 1.05M | ||||
| 2–25 | Claude Opus 5 | Anthropic | Rating 65.8 ±10.0 | $5 | $25 | 1M | Anthropic Released 2026-07 Price $5 / $25 Length 1M | |||
| 2–25 | Grok 4.6 | xAI | Rating 65.3 ±10.2 | $2 | $6 | 500K | xAI Released 2026-08 Price $2 / $6 Length 500K | |||
| 2–25 | GPT-6 Astra | OpenAI | Rating 64.7 ±11.0 | $10 | $50 | 1.05M | OpenAI Released 2026-09 Price $10 / $50 Length 1.05M | |||
| 2–25 | Gemini 3.8 Flash | Rating 64.3 ±10.2 | $0.75 | $3.75 | 1.05M | Google Released 2026-09 Price $0.75 / $3.75 Length 1.05M | ||||
| 2–27 | Claude Fable 5 | Anthropic | Rating 63.6 ±10.4 | $10 | $50 | 1M | Anthropic Released 2026-06 Price $10 / $50 Length 1M | |||
| 2–27 | Qwen3.8 Max | Alibaba | Rating 63.3 ±9.6 | $2 | $6 | 1M | Alibaba Released 2026-08 Price $2 / $6 Length 1M | |||
| 3–34 | GPT-6.1 Sol | OpenAI | Rating 60.0 ±11.0 | $2 | $10 | 1.05M | OpenAI Released 2026-09 Price $2 / $10 Length 1.05M | |||
| 4–35 | Muse Spark 1.3 | Meta | Rating 58.6 ±10.8 | $1.25 | $4.25 | 1.05M | Meta Released 2026-09 Price $1.25 / $4.25 Length 1.05M | |||
| 4–35 | GPT-5.6 Terra | OpenAI | Rating 58.2 ±10.4 | $2 | $12 | 1.05M | OpenAI Released 2026-07 Price $2 / $12 Length 1.05M | |||
| 4–37 | GLM-5.3 | Z.ai | Rating 56.6 ±10.0 | $1.4 | $4.4 | 1M | Z.ai Released 2026-08 Price $1.4 / $4.4 Length 1M | |||
| 4–37 | Grok 4.5 | xAI | Rating 56.2 ±10.6 | $2 | $6 | 500K | xAI Released 2026-07 Price $2 / $6 Length 500K | |||
| 4–37 | Claude Haiku 5.5 | Anthropic | Rating 56.1 ±10.2 | $0.1 | $0.5 | 1M | Anthropic Released 2026-10 Price $0.1 / $0.5 Length 1M | |||
| 4–38 | GPT-5.5 | OpenAI | Rating 55.1 ±10.8 | $5 | $30 | 1.05M | OpenAI Released 2026-04 Price $5 / $30 Length 1.05M | |||
| 5–38 | Grok 4.7 | xAI | Rating 54.6 ±10.0 | $2 | $6 | 500K | xAI Released 2026-09 Price $2 / $6 Length 500K | |||
| 4–38 | Claude Sonnet 5 | Anthropic | Rating 54.5 ±10.8 | $2 | $10 | 1M | Anthropic Released 2026-06 Price $2 / $10 Length 1M | |||
| 5–38 | MiMo-V2.6 Pro | Xiaomi | Rating 54.4 ±10.2 | $0.435 | $0.87 | 1.05M | Xiaomi Released 2026-09 Price $0.435 / $0.87 Length 1.05M | |||
| 4–38 | GPT-6 Sol | OpenAI | Rating 54.3 ±11.0 | $2 | $10 | 1.05M | OpenAI Released 2026-09 Price $2 / $10 Length 1.05M | |||
| 6–39 | GLM-5.3 Flash | Z.ai | Rating 52.8 ±10.4 | $0.15 | $0.5 | 1M | Z.ai Released 2026-08 Price $0.15 / $0.5 Length 1M | |||
| 6–39 | GPT-5.4 | OpenAI | Rating 52.4 ±10.4 | $2.5 | $15 | 1.05M | OpenAI Released 2026-03 Price $2.5 / $15 Length 1.05M | |||
| 6–41 | GPT-5.6 Sol Pro | OpenAI | Rating 51.4 ±10.8 | — | — | — | OpenAI Released 2026-07 Price — Length — | |||
| 10–42 | Claude Opus 4.7 | Anthropic | Rating 49.2 ±10.6 | $5 | $25 | 1M | Anthropic Released 2026-04 Price $5 / $25 Length 1M | |||
| 10–42 | Claude Opus 4.8 | Anthropic | Rating 48.9 ±10.8 | $5 | $25 | 1M | Anthropic Released 2026-05 Price $5 / $25 Length 1M | |||
| 12–44 | Kimi K3 | Moonshot | Rating 47.8 ±10.2 | $3 | $15 | 1.05M | Moonshot Released 2026-07 Price $3 / $15 Length 1.05M | |||
| 12–43 | Qwen3.8 27B | Alibaba | Rating 47.5 ±9.6 | — | — | 262.1K | Alibaba Released 2026-08 Price — Length 262.1K | |||
| 12–44 | DeepSeek-V4 Pro (2026-08) | DeepSeek | Rating 47.3 ±11.0 | $0.66 | $1.98 | 1M | DeepSeek Released 2026-08 Price $0.66 / $1.98 Length 1M | |||
| 12–44 | Gemini 3.6 Flash | Rating 46.9 ±10.2 | $0.75 | $3.75 | 1.05M | Google Released 2026-07 Price $0.75 / $3.75 Length 1.05M | ||||
| 12–45 | Claude Opus 4.6 | Anthropic | Rating 46.3 ±10.8 | $5 | $25 | 1M | Anthropic Released 2026-02 Price $5 / $25 Length 1M | |||
| 12–45 | Muse Spark 1.2 | Meta | Rating 45.4 ±9.8 | $1.25 | $4.25 | 1.05M | Meta Released 2026-08 Price $1.25 / $4.25 Length 1.05M | |||
| 12–45 | GLM-5.2 | Z.ai | Rating 45.2 ±11.2 | $1.4 | $4.4 | 1M | Z.ai Released 2026-06 Price $1.4 / $4.4 Length 1M | |||
| 13–45 | GPT-6 Luna | OpenAI | Rating 44.3 ±10.8 | $0.1 | $0.5 | 1.05M | OpenAI Released 2026-09 Price $0.1 / $0.5 Length 1.05M | |||
| 15–46 | Claude Sonnet 4.6 | Anthropic | Rating 43.0 ±10.8 | $3 | $15 | 1M | Anthropic Released 2026-02 Price $3 / $15 Length 1M | |||
| 15–46 | GPT-5.6 Luna | OpenAI | Rating 43.0 ±10.6 | $0.2 | $1.2 | 1.05M | OpenAI Released 2026-07 Price $0.2 / $1.2 Length 1.05M | |||
| 18–47 | GLM-5.1 | Z.ai | Rating 40.9 ±10.6 | $1.4 | $4.4 | 200K | Z.ai Released 2026-04 Price $1.4 / $4.4 Length 200K | |||
| 23–47 | DeepSeek-V4.1 Flash | DeepSeek | Rating 39.5 ±9.6 | — | — | 1M | DeepSeek Released 2026-09 Price — Length 1M | |||
| 25–48 | MiniMax-M3 | MiniMax | Rating 37.7 ±9.6 | $0.3 | $1.2 | 1.05M | MiniMax Released 2026-06 Price $0.3 / $1.2 Length 1.05M | |||
| 25–48 | Kimi K2.7 Code | Moonshot | Rating 37.6 ±9.4 | $0.95 | $4 | 262.1K | Moonshot Released 2026-06 Price $0.95 / $4 Length 262.1K | |||
| 26–49 | Gemini 3.1 Pro (preview) | Rating 35.1 ±9.8 | $2 | $12 | 1.05M | Google Released 2026-02 Price $2 / $12 Length 1.05M | ||||
| 29–50 | Inkling | Thinking Machines | Rating 33.8 ±9.6 | $1.87 | $4.68 | 262.1K | Thinking Machines Released 2026-07 Price $1.87 / $4.68 Length 262.1K | |||
| 28–50 | Muse Spark Open | Meta | Rating 33.4 ±10.4 | — | — | 1.04M | Meta Released 2026-10 Price — Length 1.04M | |||
| 32–50 | Muse Spark 1.1 | Meta | Rating 31.8 ±10.0 | $1.25 | $4.25 | 1M | Meta Released 2026-07 Price $1.25 / $4.25 Length 1M | |||
| 36–50 | Gemini 3.5 Flash-Lite | Rating 29.3 ±9.6 | $0.3 | $2.5 | 1.05M | Google Released 2026-07 Price $0.3 / $2.5 Length 1.05M | ||||
| 38–50 | Gemini 3.5 Flash | Rating 27.5 ±9.8 | $1.5 | $9 | 1.05M | Google Released 2026-05 Price $1.5 / $9 Length 1.05M | ||||
| 40–50 | Qwen3.5 397B A17B | Alibaba | Rating 24.9 ±9.2 | $0.6 | $3.6 | 262.1K | Alibaba Released 2026-02 Price $0.6 / $3.6 Length 262.1K | |||
| 42–50 | NVIDIA Nemotron 3 Ultra | NVIDIA | Rating 22.7 ±8.8 | — | — | 131.1K | NVIDIA Released 2026-06 Price — Length 131.1K | |||
| 43–50 | DeepSeek-V3.2 | DeepSeek | Rating 21.3 ±8.0 | — | — | 128K | DeepSeek Released 2025-12 Price — Length 128K | |||
| 51 | GPT-OSS 120B | OpenAI | Rating 4.4 ±3.9 | — | — | 128K | OpenAI Released 2025-08 Price — Length 128K |
Best success rate over time
2025-08-05Release date2026-10-15
How to read this page
- The value is the share passed on the first try. Figures published by the organizations that made the tasks, among others, collected here.
- The success rate is the share of tries that succeeded. A gap of 1 point is 1 success out of 100 tries.
- In the Supports column the icons are, from left: can use tools / can read images / can answer in a fixed format (JSON); faded means not supported or unknown.
- Open-weights models have their contents (the weights) published, so you can also run them on your own computer or server. Models without the open-weights mark are normally used through the developer's own services (apps or API); for some models the weights status is unknown.
- Names such as ChatGPT next to a model are the consumer services its developer offers. Check on the service itself whether that model is available there.
Data: Arena, Epoch AI, Swallow LLM team (CC BY 4.0), Models.dev (MIT), updated 2026-10-24, data sources