| Ornith-1.0-35B-FP8 | inferx | 262,000 | |
| Qwen3.6 35B A3B FP8 | inferx | 262,000 | |
| Qwen3.6 27B FP8 | inferx | 262,144 | |
| Qwen3.6-35B-A3B-fp8-no-thinking | inferx | 262,000 | |
| Qwen3 Coder Next FP8 | inferx | 256,144 | |
| Qwen3-Embedding-8B | inferx | 32,768 | |
| mimo-v25 | inferx | 1,000,000 | |
| Qwen3.6 35B A3B | umans-ai | 262,144 | - Input
- Input (cache write)
- Input (cached read)
- Output
|
| GLM 5.2 | umans-ai | 405,504 | - Input
- Input (cache write)
- Input (cached read)
- Output
|
| Kimi K3 | umans-ai | 1,048,576 | - Input
- Input (cached read)
- Output
|
| Kimi K2.7 Code | umans-ai | 262,144 | - Input
- Input (cache write)
- Input (cached read)
- Output
|
| DeepSeek V4 Flash | umans-ai | 1,048,576 | - Input
- Input (cached read)
- Output
|
| Umans Coder | umans-ai | 1,048,576 | - Input
- Input (cached read)
- Output
|
| Umans Flash | umans-ai | 262,144 | - Input
- Input (cached read)
- Output
|
| DeepSeek V4 Pro | umans-ai | 1,048,576 | - Input
- Input (cached read)
- Output
|
| Inkling | thinkingmachines | 1,048,576 | - Input
- Input (cached read)
- Output
|
| LucidQuery Nexus Coder | lucidquery | 250,000 | |
| AGI-01 Frontier | lucidquery | 300,000 | |
| AGI-01 Swift | lucidquery | 300,000 | |
| LucidNova RF1 100B | lucidquery | 120,000 | |
| Meta Llama 3.1 8B Instruct (GGUF) | atomic-chat | 131,072 | |
| Qwen 3.5 9B (Q4_K_M) | atomic-chat | 32,768 | |
| Qwen 3.5 9B (MLX 4-bit) | atomic-chat | 32,768 | |
| Gemma 4 E4B Instruct (MLX 4-bit) | atomic-chat | 32,768 | |
| Gemma 4 E4B Instruct (IQ4_XS) | atomic-chat | 32,768 | |