akash
akash sells inference on Liquid Inference. A buyer names one of its models as akash/<model>, or names the model and pays the lowest offer that meets the buyer's rules.
Facts
- Handle
- akash
- Website
- https://akashml.com/
- On the exchange since
- 2026-10-05
- Models
- 7
- Offers
- 7
- Regions
- us-west
- Precisions (declared)
- bf16, fp4, fp8
- Lowest price on
- 1 of 7 models
Reliability, last 3 days
- Probes answered
- 95.38%
- Hours with an answer
- 100.0%
- Median response time
- 2.4s
Checked by the exchange
3 of 7 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
The provider states nothing about how it keeps data.
Models
- GLM 5.3: declared precision fp4, us-west, $0.190 input and $4.400 output per million tokens, $0.190 cache read per million tokens, 1M tokens of context
- GPT: OSS 120B: declared precision bf16, us-west, $0.037 input and $0.187 output per million tokens, $0.037 cache read per million tokens, 131.1K tokens of context
- GPT: OSS 20B: declared precision fp4, us-west, $0.020 input and $0.100 output per million tokens, $0.020 cache read per million tokens, 131.1K tokens of context
- Kimi K3.1 Pro: declared precision fp4, us-west, $1.750 input and $10.000 output per million tokens, $0.180 cache read per million tokens, 1M tokens of context
- Llama 3.3 70B Instruct: declared precision fp8, us-west, $0.200 input and $0.520 output per million tokens, $0.100 cache read per million tokens, 131.1K tokens of context
- Qwen 3.6 35B a3b: declared precision fp8, us-west, $0.100 input and $0.900 output per million tokens, $0.050 cache read per million tokens, 262.1K tokens of context
- Qwen 3.8 27B: declared precision fp8, us-west, $0.225 input and $1.980 output per million tokens, $0.050 cache read per million tokens, 262.1K tokens of context