Capabilities
7-day heat trend
−6.9%Pricing breakdown
Typical 3:1 output-to-input mix, per 1M tokens
Estimated monthly cost by workload
Market position
- Cheaper than 43% of tracked models
- Faster than 71% of tracked models
- Efficiency rank: #706 of 1105
Best suited for
General-purpose text generation, chat, summarization and content workloads where broad capability and low cost matter most.
About KAT-Dev-FP8
KwaiPilot KAT Dev FP8 is an FP8 KAT Dev model card for lower-memory coding inference routes.
KAT-Dev-FP8 is a Text model from Kuaishou KwaiPilot (CN). HotON.ai tracks it at $0.00 per 1M input tokens and $0.00 per 1M output tokens, with a 1K-token context window, ~159 tokens/sec throughput and 98.1% availability. Its composite efficiency score is 88/100 at an estimated $0.000 per successful task.
Compare KAT-Dev-FP8
Related market news
Frequently asked questions
How much does KAT-Dev-FP8 cost per 1M tokens?+
KAT-Dev-FP8 is tracked at $0.00 per 1M input tokens and $0.00 per 1M output tokens. A typical 3:1 output-to-input workload blends to roughly $0.00 per 1M tokens. Figures are illustrative demo data.
What is KAT-Dev-FP8 best for?+
General-purpose text generation, chat, summarization and content workloads where broad capability and low cost matter most.
How fast is KAT-Dev-FP8?+
KAT-Dev-FP8 delivers about 159 tokens/sec with 98.1% tracked availability, suitable for latency-sensitive, real-time applications.
Is KAT-Dev-FP8 cheaper than other AI models?+
Within the HotON.ai tracked set, KAT-Dev-FP8 is cheaper than 43% of models on input price and ranks #706 of 1105 by overall efficiency.
Related models
Pricing is real (via the TestKey catalog, updated daily). Quality (Arena Elo) is real where the model is ranked on LMArena. Speed, availability and efficiency are modeled estimates.