| Qwen3-Coder 30B-A3B Instruct | pendra | 262,144 | |
| Qwen3.6 27B | pendra | 262,144 | |
| Llama-3.3-70B-Instruct | pendra | 128,000 | |
| GLM-4.7-Flash | pendra | 203,000 | - Input
- Input (cache write)
- Input (cached read)
- Output
|
| Gemma 4 26B A4B IT | scaleway | 262,000 | - Input
- Input (cached read)
- Output
|
| BGE Multilingual Gemma2 | scaleway | 8,191 | |
| Qwen2.5-Omni 7B | qwen | 32,768 | |
| Qwen-MT Turbo | qwen | 16,384 | |
| Qwen Deep Research | qwen | 1,000,000 | |
| Moonshot Kimi K2 Instruct | moonshotai | 131,072 | |
| Qwen-Omni Turbo Realtime | qwen | 32,768 | - Input
- Input (audio)
- Output
- Output (audio)
|
| Qwen3-VL 30B-A3B | qwen | 131,072 | - Input
- Output
- Output (reasoning)
|
| Qwen-VL OCR | qwen | 34,096 | |
| Tongyi Intent Detect V3 | qwen | 8,192 | |
| Qwen2.5-Math 7B Instruct | qwen | 4,096 | |
| Qwen3-ASR Flash | qwen | 53,248 | |
| Qwen3-Omni Flash | qwen | 65,536 | - Input
- Input (audio)
- Output
- Output (audio)
|
| Qwen2.5-Math 72B Instruct | qwen | 4,096 | |
| Qwen3-Omni Flash Realtime | qwen | 65,536 | - Input
- Input (audio)
- Output
- Output (audio)
|
| Qwen2.5 14B Instruct | qwen | 131,072 | |
| Qwen Math Turbo | qwen | 4,096 | |
| Qwen Plus Character | qwen | 32,768 | |
| Qwen2.5 32B Instruct | qwen | 131,072 | |
| Qwen3-VL 235B-A22B | qwen | 256,000 | - Input
- Input (cached read)
- Output
|
| Qwen Math Plus | qwen | 4,096 | |