All documented models
10 public models with source-linked specifications, provider pricing, and benchmark records.
Ranked by raw score for SWE-Atlas Test Writing · v. Versions are kept separate.
| Model | Creator | Raw benchmark score | Modalities | Context | Input / 1M | Output / 1M | Released | |
|---|---|---|---|---|---|---|---|---|
| GPT-5.4openai/gpt-5.4 |  OpenAI | 44.36 | Image, Pdf, Text | 1.05M | $2.500 | $15.000 | 2026-03-05 | |
| GPT-5.5openai/gpt-5.5 | OpenAI | 42.59 | Image, Pdf, Text | 1.05M | $5.000 | $30.000 | 2026-04-23 | |
| GPT-5.3 Codexopenai/gpt-5.3-codex | OpenAI | 38.98 | Image, Pdf, Text | 400K | $1.750 | $14.000 | 2026-02-05 | |
| Claude Opus 4.6anthropic/claude-opus-4-6 |  Anthropic | 36.67 | Image, Pdf, Text | 1M | $5.000 | $25.000 | 2026-02-05 | |
| Claude Sonnet 4.6anthropic/claude-sonnet-4-6 | Anthropic | 31.76 | Image, Pdf, Text | 1M | $3.000 | $15.000 | 2026-02-17 | |
| Gemini 3 Flash Previewgoogle/gemini-3-flash-preview |  Google | 30.3 | Audio, Image, Pdf, Text, Video | 1.04858M | $0.500 | $3.000 | 2025-12-17 | |
| Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview | Google | 29.84 | Audio, Image, Pdf, Text, Video | 1.04858M | $2.000 | $12.000 | 2026-02-19 | |
| GLM-5zhipuai/glm-5 |  Zhipu AI | 28.74 | Text | 204.8K | $1.000 | $3.200 | 2026-02-12 | |
| Kimi K2.5moonshotai/kimi-k2.5 |  Moonshot AI | 25.77 | Image, Text, Video | 262.144K | $0.600 | $3.000 | 2026-01 | |
| MiniMax-M2.5minimax/MiniMax-M2.5 |  MiniMax | 18.6 | Text | 204.8K | $0.300 | $1.200 | 2026-02-12 |
Page 1 of 1