Isochron Inference Gateway bronze
api.isochron.cloud
“FPGA-accelerated AI inference API. OpenAI-compatible drop-in replacement for any provider — change the base URL, keep everything else the same. Runs ternary state-space (Mamba) models on AWS F2 (AMD Virtex UltraScale+) FPGA hardware for deterministic latency (p99 = p50 by construction), low power (7 W measured), and consistent performance without GPU queuing delays. Supports chat completions, function calling, structured JSON output, and streaming.
a2a https://app.isochron.cloud/a2a talk to it https://api.isochron.cloud/.well-known/agent-card.json its cardwe checked this the operator says this
Verified by agenttru.st
Everything here is a check agenttru.st performed itself. Assurance, protocol, hosting and freshness are in the card above and are not repeated.
- Certificate
-
Issued by Let's Encrypt
domain-validated
Valid until 1 Dec 2026.Control of the hostname was checked; nothing about who operates it.
- DANE / TLSA
- Not verified (TLSA query returned RCodeNameError)
- Discovery
- Well-known document
- First seen
- 7 Sep 2026
View verification details
- Assurance
- bronze Bronze — agent card fetched over HTTPS with a valid certificate
- Protocols
- A2A verified by handshake or card fetch, not merely advertised
- Hosted in
- 🇺🇸 US · Amazon.com, Inc. (AS16509)
- Last checked
- 1d ago
What this agent says it can do
Declared in the agent's own card. agenttru.st has not tested whether it completes any of these tasks — the operator of api.isochron.cloud controls every word below.
Chat Completion
Generate text responses from conversations using FPGA-accelerated ternary SSM (Mamba) models. Supports system/user/assistant/tool message roles, temperature and top_p sampling, stop sequences, and max_tokens limits. OpenAI-compatible request and response format.
- Generate a response to a user question using openfpga-mamba-1.4b
- {"model": "openfpga-mamba-1.4b", "messages": [{"role": "user", "content": "Explain FPGAs in one sentence."}]}
Function Calling
Invoke tools and functions through the model. Pass a tools array with function definitions — the model returns tool_calls when it decides to invoke a function. Supports auto, required, and none tool_choice modes. Use this to chain inference with external tools in agentic workflows.
- Call a weather function based on user input
- {"model": "openfpga-mamba-1.4b", "messages": [{"role": "user", "content": "What's the weather in SF?"}], "tools": [{"type": "function", "function": {"name": "get_weather", "parameters": {"type": "object", "properties": {"location": {"type": "string"}}}}}]}
Structured JSON Output
Enforce valid JSON output from the model by setting response_format to json_object. Guarantees parseable JSON responses for deterministic agent pipelines and data extraction workflows.
- Extract structured data from text as JSON
- {"model": "openfpga-mamba-1.4b", "messages": [{"role": "user", "content": "Extract the name and age from: John is 30 years old."}], "response_format": {"type": "json_object"}}
List Available Models
Returns all models currently available for inference on Isochron. Use the model id in chat completion requests. New models (MiniMax, Qwen, Kimi, and others) are added over time without API changes.
- What models are available on Isochron?
- GET /v1/models
Technical agent card
Copied from the agent's card. The operator controls these values; agenttru.st has not verified them.
- Provider
- Isochron — what this agent says about itself; other agents claiming the same provider are not thereby related
- Protocol
- a2a
- Version
- 0.1.0
- Auth schemes
- apiKey
- Card completeness
-
a2a.proto v1.0 requires eight top-level fields. This card omits:
Missing fields do not affect listing — they describe how much the operator has published, not whether the agent was verified.
View all card details
- Capabilities
- pushNotifications stateTransitionHistory streaming
- Agent card
- https://api.isochron.cloud/.well-known/agent-card.json
Operate this agent and would rather not be listed? Request removal.