SOKKAN
Swiss provider (Geneva). Zero data retention, no training, on every offer. Own GPUs in Geneva, dedicated rented H100 in the US, OVHcloud in the EU.
SOKKAN sells inference on Liquid Inference. A buyer names one of its models as sokkan/<model>, or names the model and pays the lowest offer that meets the buyer's rules.
Facts
- Handle
- sokkan
- Website
- https://sokkan.ch/
- Terms of service
- https://sokkan.ch/en/api/
- Privacy policy
- https://sokkan.ch/en/privacy/
- On the exchange since
- 2026-10-28
- Models
- 8
- Offers
- 8
- Regions
- eu
- Precisions (declared)
- fp8, provider-managed
- Lowest price on
- 1 of 8 models
Reliability, last 3 days
- Probes answered
- 95.39%
- Hours with an answer
- 90.6%
- Median response time
- 2.2s
Checked by the exchange
2 of 8 offers match the fingerprint of the model. 0 have a finished evaluation.
The exchange sends fixed prompts to each offer and compares the answers with a baseline for the model. It also runs benchmarks against the offers and keeps the results.
Data handling
- Zero retention, declared by the provider
- No training, declared by the provider
- GDPR DPA, declared by the provider
Models
- DeepSeek V4.1 Flash 1017: declared precision provider-managed, eu, $0.690 input and $2.050 output per million tokens, $0.140 cache read per million tokens, 262.1K tokens of context
- GLM 5.3: declared precision provider-managed, eu, $2.050 input and $6.200 output per million tokens, $2.050 cache read per million tokens, 200K tokens of context
- GPT: OSS 120B: declared precision provider-managed, eu, $0.090 input and $0.460 output per million tokens, $0.090 cache read per million tokens, 131.1K tokens of context
- Mistral: Small 3.2 24B Instruct: declared precision provider-managed, eu, $0.100 input and $0.320 output per million tokens, $0.100 cache read per million tokens, 128K tokens of context
- Qwen 3.5 397B a17b: declared precision provider-managed, eu, $0.168 input and $1.158 output per million tokens, $0.168 cache read per million tokens, 131.1K tokens of context
- Qwen 3.6 27B: declared precision provider-managed, eu, $0.164 input and $1.200 output per million tokens, $0.164 cache read per million tokens, 131.1K tokens of context
- Qwen 3.6 35B a3b: declared precision fp8, eu, $0.043 input and $0.180 output per million tokens, $0.043 cache read per million tokens, 262.1K tokens of context
- Qwen 3.8 27B: declared precision fp8, eu, $0.024 input and $0.180 output per million tokens, $0.024 cache read per million tokens, 131.1K tokens of context