Counts come from the response
The tested stream finishes with usage.prompt_tokens, usage.completion_tokens and their total. These are model-side counts, not character-count estimates.
Every estimate distinguishes the tokens you send from the tokens the model generates. Rates vary by model because model size, architecture and serving requirements affect the cost of inference.
| Model | Details | Input / 1M | Output / 1M | Availability |
|---|---|---|---|---|
| MiniCPM · 1B Uncensored hf/dealignai/MiniCPM-V-4.7-1B-CRACK-GGUF minicpm-1b-uncensored A small text model used to verify the actual streaming and token-usage path on bounded CPU infrastructure. | GGUF · Q4_K_M1BUncensored | $0.10USD / preview | $0.30USD / preview | Integration tested |
| Qwen3 · 8B Abliterated v2 hf/huihui-ai/Huihui-Qwen3-8B-abliterated-v2 qwen3-8b-abliterated Qwen3-8B Abliterated v2, with a 16K-context serving configuration on one L4 GPU. | SafeTensors · BF168BAbliterated | $0.18USD / preview | $0.18USD / preview | Integration tested |
| Qwen3 · 32B Abliterated hf/huihui-ai/Qwen3-32B-abliterated qwen3-32b-abliterated A larger abliterated model for longer generations and more involving text tasks. | SafeTensors · BF1632BAbliterated | $0.75USD / preview | $0.75USD / preview | Planned |
| GLM 5.3 Flash · Uncensored hf/dealignai/GLM-5.3-Flash-UNCENSORED-FP8 glm-5.3-flash-uncensored A large but more efficient general-purpose variant, with reasoning and a vision tower reported by the source card. | SafeTensors · FP8320B reportedUncensored | $1.00USD / preview | $3.00USD / preview | Planned |
| GLM 5.3 · Uncensored hf/dealignai/GLM-5.3-UNCENSORED-FP8 glm-5.3-uncensored The full-size general-purpose GLM release for demanding text and reasoning workloads. | SafeTensors · FP8753B reportedUncensored | $2.00USD / preview | $8.00USD / preview | Planned |
| DeepSeek V4.1 Flash 1017 · Uncensored hf/dealignai/DeepSeek-V4.1-Flash-1017-UNCENSORED-FP8 deepseek-v4.1-flash-1017-uncensored A long-context multimodal research model with reasoning tiers and native tool support reported upstream. | SafeTensors · FP8 + FP4763B reportedUncensored | $1.20USD / preview | $4.00USD / preview | Planned |
| Kimi K3.1 Pro · Uncensored hf/lyber-ai/Kimi-K3.1-Pro-UNCENSORED-MXFP4 kimi-k3.1-pro-uncensored An extended-reasoning release with reported native tools and 128K+ source context. Upstream access approval is required. | SafeTensors · MXFP4128K+ source contextUncensored | $5.00USD / preview | $24.00USD / preview | Planned |
Only “Integration tested” entries are exposed through the configured inference route. Planned entries may show source-reported capabilities; they are not Blaschke serving guarantees or enabled API model IDs.
Enter whole numbers for each token count. The result uses the displayed preview rates; it is not a measured serving cost or a customer charge.
10,000 input + 2,000 output tokens
Input cost = prompt_tokens × input_rate_per_1M ÷ 1,000,000
Output cost = completion_tokens × output_rate_per_1M ÷ 1,000,000
Estimated total = input cost + output cost
Example using qwen3-8b-abliterated:
12,000 prompt tokens, 2,000 completion tokens
= (12,000 × $0.18 + 2,000 × $0.18) ÷ 1,000,000
= $0.001800
The console reports usage from the model; these prices are preview-only.The tested stream finishes with usage.prompt_tokens, usage.completion_tokens and their total. These are model-side counts, not character-count estimates.
Tokens you send are priced at the model’s input rate. Tokens generated are priced at its output rate. An empty output shows zero reported completion tokens; a terminated generation can have zero too.
Displaying these schedules does not activate a payment provider or charge a user. Before commercial launch, any minimum per-request contribution or output reservation will need to be documented explicitly.
Large-model prices anticipate the family’s likely compute demands rather than measured serving costs. Hardware selection, batching and sustained utilisation could lead to a different final rate.
This describes the prototype preview schedule. There is no paid account, no prepaid wallet and no uptime commitment yet. Caching is not offered or priced as a discount. Tool execution is not included as part of the token rate.