Mancer is a small service that hosts language models behind an API, aimed at roleplay and creative writing use.
- Type
- Private company
- AI lab
- Inference provider
- 10 hosted endpoints
| Model | Endpoint | Context | Input /M | Output /M | Cached input /M | Cache write /M | Uptime 1d |
|---|---|---|---|---|---|---|---|
 DeepSeek V4 Flash 0423 | mancer/fp8 | 1M | $0.15 | $0.50 | $0.075 | — | 100.00% |
| DeepSeek V4 Flash 0731 | mancer/fp8 | 1M | $0.15 | $0.50 | $0.075 | — | 99.90% |
| DeepSeek V4.1 Flash | mancer/fp8 | 1M | $0.2125 | $0.75 | $0.05 | — | 99.92% |
 GLM 4.7 | mancer/fp4 | 131K | $0.70 | $2.50 | $0.35 | — | 99.68% |
| Magnum v4 72B | mancer/fp8 | 33K | $2.50 | $5.00 | $2.50 | — | 100.00% |
| MythoMax 13B | mancer/fp8 | 8K | $0.30 | $0.50 | $0.20 | — | 100.00% |
 Qwen3.8 27B | mancer/fp8 | 262K | $0.15 | $1.25 | $0.05 | — | 100.00% |
| ReMM SLERP 13B | mancer/fp8 | 6K | $0.30 | $0.50 | $0.20 | — | 99.48% |
 Weaver (alpha) | mancer/fp8 | 8K | $0.30 | $0.50 | $0.30 | — | 100.00% |
 gpt-oss-120b | mancer/fp8 | 131K | $0.055 | $0.25 | $0.055 | — | 100.00% |