| Rank | Model | Provider | Score | Benchmarks | Inference | Agentic | Programming | Value | Price |
|---|---|---|---|---|---|---|---|---|---|
| 21 | Muse Spark 1.1 muse-spark-1.1 multimodalvisionmulti-input reasoning | Meta | 65.8 overall | 66.4 | 81.9 | 75.6 | 55.1 | 38.8 | $1.25 in / $4.25 out |
| 22 | Seed 2.1 Pro seed-2.1-pro multimodalvisionmulti-input reasoning | ByteDance | 65.4 overall | 67.2 | 0.0 | 68.4 | 59.9 | 0.0 | N/A |
| 23 | GLM-5.3 glm-5.3 textinference | Zhipu AI | 64.7 overall | 68.8 | 81.9 | 59.9 | 0.0 | 37.4 | $1.4 in / $4.4 out |
| 24 | Qwen3.8 Flash qwen3.8-flash multimodalvisionmulti-input reasoning | Alibaba Cloud / Qwen Team | 63.5 overall | 61.1 | 58.6 | 64.5 | 60.0 | 83.5 | $0.15 in / $0.47 out |
| 25 | Claude Fable 5 claude-fable-5 multimodalvisionmulti-input reasoning | Anthropic | 62.5 overall | 69.5 | 58.6 | 0.0 | 84.2 | 1.9 | |
| 26 | Qwen3.8-Flash-Next qwen3.8-flash-next multimodalvisionmulti-input reasoning | Alibaba Cloud / Qwen Team | 61.9 overall | 61.1 | 0.0 | 64.5 | 60.0 | 0.0 | N/A |
| 27 | MiMo-V2-Pro mimo-v2-pro codeprogrammingtool use | Xiaomi | 61.7 overall | 0.0 | 0.0 | 0.0 | 61.7 | 0.0 | N/A |
| 28 | GPT-5.1 Codex High gpt-5.1-codex-high multimodalvisionmulti-input reasoning | OpenAI | 61.4 overall | 61.4 | 0.0 | 0.0 | 0.0 | 0.0 | |
| 29 | GPT-5 High gpt-5-high-2025-08-07 multimodalvisionmulti-input reasoning | OpenAI | 60.8 overall | 60.8 | 0.0 | 0.0 | 0.0 | 0.0 | |
| 30 | GPT-5.5 gpt-5.5 multimodalvisionmulti-input reasoning | OpenAI | 60.6 overall | 74.2 | 95.2 | 54.9 | 49.2 | 5.8 | $5 in / $30 out |
| 31 | Claude Opus 4.8 claude-opus-4-8 multimodalvisionmulti-input reasoning | Anthropic | 60.0 overall | 72.5 | 26.7 | 68.7 | 81.8 | 9.5 | |
| 32 | Claude Opus 5 claude-opus-5 multimodalvisionmulti-input reasoning | Anthropic | 59.9 overall | 70.5 | 58.6 | 69.4 | 0.0 | 9.2 | |
| 33 | DeepSeek-V3.2 (Non-thinking) deepseek-chat textinference | DeepSeek | 59.7 overall | 0.0 | 48.7 | 0.0 | 0.0 | 77.3 | $0.28 in / $0.42 out |
| 34 | Qwen3.7 Max qwen3.7-max multimodalvisionmulti-input reasoning | Alibaba Cloud / Qwen Team | 59.3 overall | 64.0 | 58.6 | 48.0 | 74.4 | 40.8 | $1.25 in / $3.75 out |
| 35 | Seed 2.1 Turbo seed-2.1-turbo multimodalvisionmulti-input reasoning | ByteDance | 59.1 overall | 63.6 | 0.0 | 57.6 | 55.1 | 0.0 | N/A |
| 36 | Kimi K2-Thinking-0905 kimi-k2-thinking-0905 codeprogrammingtool use | Moonshot AI | 59.0 overall | 64.9 | 0.0 | 51.2 | 59.9 | 0.0 | |
| 37 | GPT-5.5 Pro gpt-5.5-pro multimodalvisionmulti-input reasoning | OpenAI | 58.9 overall | 59.9 | 0.0 | 67.1 | 48.8 | 0.0 | N/A |
| 38 | GLM-5.2 glm-5.2 codeprogrammingtool use | Zhipu AI | 58.7 overall | 66.8 | 81.9 | 39.1 | 57.9 | 47.6 | $0.95 in / $3 out |
| 39 | Gemini 3.8 Flash gemini-3.8-flash multimodalvisionmulti-input reasoning | Google | 58.5 overall | 50.6 | 81.9 | 0.0 | 0.0 | 43.2 | |
| 40 | Gemini 3 Pro gemini-3-pro-preview multimodalvisionmulti-input reasoning | Google | 58.3 overall | 68.8 | 0.0 | 51.0 | 52.9 | 0.0 |
21
Muse Spark 1.1
Meta
65.8
$1.25 in / $4.25 out
22
Seed 2.1 Pro
ByteDance
65.4
N/A
23
GLM-5.3
Zhipu AI
64.7
$1.4 in / $4.4 out
24