| Model | Context | Input | Output | Throughput | Latency | Providers | Agents | Coding | Reasoning | Knowledge | Math | Multilingual | Multimodal | Instruction following |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| GLM-3 Turbo | — | — | — | — | — | +19 | — | — | — | — | — | — | — | — |
| Kimi K2 Instruct | — | — | — | 129 t/s | 1.05s | +76 | — | — | — | — | — | — | — | — |
| GPT-4o Mini Realtime | — | — | — | — | — | +30 | — | — | — | — | — | — | — | — |
| MiniMax M2.7 HighSpeed | — | — | — | 49 t/s | 3.53s | +48 | — | — | — | — | — | — | — | — |
| Grok 3 | — | — | — | 43 t/s | 2.64s | +82 | — | 39.7 | 51.4 | — | 60 | — | — | — |
| Claude 1.3 | — | — | — | — | — | +1 | — | — | — | — | — | — | — | — |
| Qwen3 TTS Flash | — | — | — | — | — | +26 | — | — | — | — | — | — | — | — |
| GPT-4o Realtime | — | — | — | — | — | +63 | — | — | — | — | — | — | — | — |
| Gemini 2.0 Flash 001 | — | — | — | 108 t/s | 2.70s | +20 | — | — | — | — | — | — | — | — |
| BGE Reranker V2 M3 | — | — | — | — | — | +36 | — | — | — | — | — | — | — | — |
| Gemini 2.5 Flash Native Audio | — | — | — | — | — | +5 | — | — | — | — | — | — | — | — |
| GLM-4V | — | — | — | — | — | +17 | — | — | — | — | — | — | — | — |
| GLM-Z1 Flash | — | — | — | 109 t/s | 0.45s | +18 | — | — | — | — | — | — | — | — |
| Text Embedding 3 Small | — | — | — | — | — | +48 | — | — | — | — | — | — | — | — |
| GLM-Z1 AirX | — | — | — | 168 t/s | 0.34s | +7 | — | — | — | — | — | — | — | — |
| GPT-4o Transcribe Diarize | — | — | — | — | — | +16 | — | — | — | — | — | — | — | — |
| FLUX Kontext Pro | — | — | — | — | — | +16 | — | — | — | — | — | — | — | — |
| ChatGPT Image | — | — | — | — | — | +11 | — | — | — | — | — | — | — | — |
| GPT-5 Codex Mini | — | — | — | 38 t/s | 3.35s | +60 | — | — | — | — | — | — | — | — |
| GPT-4o Image | — | — | — | — | — | +27 | — | — | — | — | — | — | — | — |
| MiniMax M2.5 Lightning | — | — | — | — | — | +6 | — | — | — | — | — | — | — | — |
| GPT-4 Net | — | — | — | — | — | +2 | — | — | — | — | — | — | — | — |
| GPT Image 1.5 | — | — | — | — | — | +97 | — | — | — | — | — | — | — | — |
| Qwen3 TTS Flash Realtime | — | — | — | — | — | +13 | — | — | — | — | — | — | — | — |
| GPT-4 Gizmo | — | — | — | — | — | +23 | — | — | — | — | — | — | — | — |
| MiniMax Feed | — | — | — | — | — | +1 | — | — | — | — | — | — | — | — |
| Net GPT-4o | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GPT-4o Mini Transcribe | — | — | — | — | — | +33 | — | — | — | — | — | — | — | — |
| MiniMax-Text-01 | — | — | — | — | — | +2 | — | — | — | — | — | — | — | — |
| GLM-4 Long | — | — | — | — | — | +15 | — | — | — | — | — | — | — | — |
| Claude Instant 1.2 | — | — | — | — | — | +6 | — | — | — | — | — | — | — | — |
| ChatGPT-4o | — | — | — | 78 t/s | 4.56s | +44 | — | — | — | — | — | — | — | — |
| Grok 3 Mini | — | — | — | 122 t/s | 5.23s | +68 | — | — | — | — | — | — | — | — |
| Qwen3.5 Plus Image | — | — | — | — | — | +1 | — | — | — | — | — | — | — | — |
| Claude 2.1 | — | — | — | — | — | +8 | — | 19 | 28.5 | — | 20.4 | — | — | — |
| GPT-4o Mini TTS | — | — | — | — | — | +35 | — | — | — | — | — | — | — | — |
| GLM-4.5 AirX | — | — | — | 71 t/s | 11.56s | +21 | — | — | — | — | — | — | — | — |
| Gemini Flash Lite | — | — | — | — | — | +36 | — | — | — | — | — | — | — | — |
| Text Embedding 3 Large | — | — | — | — | — | +49 | — | — | — | — | — | — | — | — |
| GPT-4 Vision | — | — | — | — | — | +29 | — | — | — | — | — | — | — | — |
| Gemini 1.5 Pro 001 | — | — | — | — | — | +1 | — | — | — | — | — | — | — | — |
| GPT-SoVITS | — | — | — | — | — | +1 | — | — | — | — | — | — | — | — |
| Qwen3 S2S Flash Realtime | — | — | — | — | — | +12 | — | — | — | — | — | — | — | — |
| GPT-4.5 | — | — | — | — | — | +34 | — | — | — | — | — | — | — | — |
| Claude 3.5 Sonnet | — | — | — | 21 t/s | 3.36s | +24 | — | 41.2 | 47 | — | 46.4 | — | — | — |
| DeepSeek Reasoner | — | — | — | 30 t/s | 7.02s | +79 | — | — | — | — | — | — | — | — |
| Qwen3 Omni Flash | — | — | — | 62 t/s | 5.13s | +34 | — | — | — | — | — | — | — | — |
| Devstral Small | — | — | — | — | — | +6 | — | 24.9 | 35.8 | — | 31.9 | — | — | — |
| Gemini 1.5 Pro | — | — | — | 18 t/s | 2.58s | +20 | — | 28.6 | 46.3 | 24.6 | 55.3 | — | 55 | — |
| Gemini 3 Pro | — | — | — | 73 t/s | 10.59s | +123 | 75.2 | 49.5 | 52.8 | 55 | — | — | 82.5 | 70.4 |
| O1 Mini | — | — | — | 32 t/s | 13.84s | +52 | — | 45 | 46.5 | — | 77.4 | — | — | — |
| MiniMax Hailuo 2.3 | — | — | — | — | — | +11 | — | — | — | — | — | — | — | — |
| Qwen3.6 Plus | 1M | — | — | 47 t/s | 24.53s | +110 | 56 | 55.7 | 57.9 | 52.6 | 91.6 | 71.3 | 84.6 | 81.8 |
| GLM-5V Turbo | 202.8K | — | — | — | — | +33 | 61 | 39.9 | 39.6 | 45.6 | — | — | 86.4 | 61.1 |
| Grok 4.20 | 2M | — | — | 75 t/s | 4.82s | +84 | 49.4 | 50.5 | 58.9 | 47.7 | — | — | 73.3 | — |
| MiniMax M2.7 | 204.8K | — | — | 178 t/s | 7.32s | +150 | 52.3 | 53.3 | 46.2 | 47.8 | 80 | — | 100 | 75.7 |
| GPT-5.4 Nano | 400K | — | — | 137 t/s | 3.38s | +83 | 48.4 | 43.7 | 45.9 | 45 | — | — | 67 | 75.9 |
| GPT-5.4 Mini | 400K | — | — | 138 t/s | 3.62s | +210 | 56.4 | 46.2 | 44.7 | 51.6 | — | — | 76 | 73.3 |
| GLM-5 Turbo | 262.1K | — | — | 56 t/s | 15.05s | +69 | 77.2 | 40.2 | 42.8 | 47.9 | — | — | 100 | 73.2 |
| Qwen3.5 | 262.1K | — | — | 48 t/s | 10.60s | +86 | 51 | 51.1 | 53.1 | 66.1 | 89.9 | 71.9 | 81.7 | 85.7 |