Running @dphnAI API on my local machine using DeepSeek harness. 99% Cache hit, 0.2$ cost to use 16M tokens. ๐
base:0xed664536023d8e4b1640c394777d34abaff1df8f
$POD launches inference API powered by DeepSeek V4.1 Flash.
@dphnAI is an AI lab that makes unrestricted/open models and runs a distributed inference network. Dolphin models have millions of downloads on Hugging Face.
A mix of idle consumer GPUs and data center GPUs contributes compute to the network. Dolphin routes inference requests on the network.
The network has so far processed ~8T+ tokens across ~1.6k unique workers.
The API signals Dolphin's expansion from building models and the network to a full inference API for agents, which is cheap, high-performance, and optimized for agentic workloads.
Dolphin has committed 100% of protocol revenue to open-market $POD token buybacks. However, multiple independent analyses confirm that no token buybacks have occurred to date.
API usage revenue could finally kick off the buyback flywheel, driving sustained programmatic demand for $POD.
$POD launches inference API powered by DeepSeek V4.1 Flash.
@dphnAI is an AI lab that makes unrestricted/open models and runs a distributed inference network. Dolphin models have millions of downloads on Hugging Face.
A mix of idle consumer GPUs and data center GPUs contributes compute to the network. Dolphin routes inference requests on the network.
The network has so far processed ~8T+ tokens across ~1.6k unique workers.
The API signals Dolphin's expansion from building models and the network to a full inference API for agents, which is cheap, high-performance, and optimized for agentic workloads.
Dolphin has committed 100% of protocol revenue to open-market $POD token buybacks. However, multiple independent analyses confirm that no token buybacks have occurred to date.
API usage revenue could finally kick off the buyback flywheel, driving sustained programmatic demand for $POD.
@CryptoWizardd@KAPOTHEGOAT01 Both of yall are legends. Love to have both of yall in base:0xed664536023d8e4b1640c394777d34abaff1df8f with @jkrdoc and us
$POD launches inference API powered by DeepSeek V4.1 Flash.
@dphnAI is an AI lab that makes unrestricted/open models and runs a distributed inference network. Dolphin models have millions of downloads on Hugging Face.
A mix of idle consumer GPUs and data center GPUs contributes compute to the network. Dolphin routes inference requests on the network.
The network has so far processed ~8T+ tokens across ~1.6k unique workers.
The API signals Dolphin's expansion from building models and the network to a full inference API for agents, which is cheap, high-performance, and optimized for agentic workloads.
Dolphin has committed 100% of protocol revenue to open-market $POD token buybacks. However, multiple independent analyses confirm that no token buybacks have occurred to date.
API usage revenue could finally kick off the buyback flywheel, driving sustained programmatic demand for $POD.
$POD is up ~22.2% in the last 24 hours.
Running @dphnAI API on my local machine using DeepSeek harness. 99% Cache hit, 0.2$ cost to use 16M tokens. ๐
base:0xed664536023d8e4b1640c394777d34abaff1df8f
Running @dphnAI API on my local machine using DeepSeek harness. 99% Cache hit, 0.2$ cost to use 16M tokens. ๐
base:0xed664536023d8e4b1640c394777d34abaff1df8f
Running @dphnAI API on my local machine using DeepSeek harness. 99% Cache hit, 0.2$ cost to use 16M tokens. ๐
base:0xed664536023d8e4b1640c394777d34abaff1df8f
Running @dphnAI API on my local machine using DeepSeek harness. 99% Cache hit, 0.2$ cost to use 16M tokens. ๐
base:0xed664536023d8e4b1640c394777d34abaff1df8f
Running @dphnAI API on my local machine using DeepSeek harness. 99% Cache hit, 0.2$ cost to use 16M tokens. ๐
base:0xed664536023d8e4b1640c394777d34abaff1df8f
Running @dphnAI API on my local machine using DeepSeek harness. 99% Cache hit, 0.2$ cost to use 16M tokens. ๐
base:0xed664536023d8e4b1640c394777d34abaff1df8f
$POD launches inference API powered by DeepSeek V4.1 Flash.
@dphnAI is an AI lab that makes unrestricted/open models and runs a distributed inference network. Dolphin models have millions of downloads on Hugging Face.
A mix of idle consumer GPUs and data center GPUs contributes compute to the network. Dolphin routes inference requests on the network.
The network has so far processed ~8T+ tokens across ~1.6k unique workers.
The API signals Dolphin's expansion from building models and the network to a full inference API for agents, which is cheap, high-performance, and optimized for agentic workloads.
Dolphin has committed 100% of protocol revenue to open-market $POD token buybacks. However, multiple independent analyses confirm that no token buybacks have occurred to date.
API usage revenue could finally kick off the buyback flywheel, driving sustained programmatic demand for $POD.
$POD is up ~22.2% in the last 24 hours.