Documentation
Ophiris API
Ophiris serves 58 uncensored, abliterated and security-tuned models through one OpenAI-compatible API, with per-key spending limits and private billing records. Any client that speaks Chat Completions works with a base URL and a key.
Quickstart
- Open the app and sign in with Google, X, GitHub or an Ethereum wallet. Signing in sends no transaction; deposits, Pay with OPHIRIS and holder credits use a wallet you link on Robinhood Chain.
- In API access, create a key. Choose the models it may use and its spending limit. The key is shown once.
- Call the API at
https://www.ophiris.xyz/api/v1withAuthorization: Bearer oph_sk_….
curl https://www.ophiris.xyz/api/v1/chat/completions \ -H "Authorization: Bearer $OPHIRIS_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "google/gemma-4-26b-a4b-it-cybersecurity", "stream": true, "stream_options": { "include_usage": true }, "messages": [{ "role": "user", "content": "Write a haiku about latency." }] }'import osfrom openai import OpenAI client = OpenAI(base_url="https://www.ophiris.xyz/api/v1", api_key=os.environ["OPHIRIS_API_KEY"])reply = client.chat.completions.create( model="google/gemma-4-26b-a4b-it-cybersecurity", messages=[{"role": "user", "content": "Summarize this changelog."}],)print(reply.choices[0].message.content)Use www.ophiris.xyz. Requests to ophiris.xyz are redirected, and most HTTP clients drop the Authorization header when they follow a redirect.
Try it
Create a demo API key without an account, then call the API with it here, with curl or from any OpenAI-compatible client. Demo keys are ordinary Ophiris keys on a shared demo account: every request is held and settled in the billing ledger, and the key's spending limit and model list apply. Each key covers a few efficient models, at most 1,024 output tokens per request and no web tools, and expires after 24 hours. Each visitor can create 3 keys every 24 hours.
Loading the demo…
curl https://www.ophiris.xyz/api/demo # models and limits of demo keyscurl -X POST https://www.ophiris.xyz/api/demo/keys # returns { "key": "oph_sk_…", "models": [...], "expires_at": … } curl https://www.ophiris.xyz/api/v1/chat/completions \ -H "Authorization: Bearer oph_sk_…" \ -H "Content-Type: application/json" \ -d '{ "model": "<a model the key allows>", "messages": [{ "role": "user", "content": "Name three uses of embeddings." }] }'Models
The catalog lists each model's provider, context window, maximum output, capabilities and price per million input and output tokens. Browse it on Models, or list it in OpenAI format. Without a key the list is public; with a key it is limited to the models that key may use.
curl https://www.ophiris.xyz/api/v1/models{ "id": "google/gemma-4-26b-a4b-it-cybersecurity", "object": "model", "owned_by": "…", "context_length": 128000, "max_output_tokens": 16384, "capabilities": { "tools": true, "reasoning": false, "vision": true, "structured_output": true }, "reasoning_efforts": [], "pricing": { "prompt": 0.15, "completion": 0.6, "cache_read": null, "cache_write": null, "unit": "usd_per_million_tokens" }}Chat completions
POST https://www.ophiris.xyz/api/v1/chat/completions implements OpenAI Chat Completions, streaming (server-sent events) or not.
- Messages:
system,developer,user,assistantandtoolroles; assistanttool_calls; inline PNG, JPEG, WebP and GIF images as data URLs for vision models;cache_controlhints. Message content is forwarded byte for byte, so prompt caches keep their prefixes. - Tools:
tools,tool_choiceandparallel_tool_callson models with tool calling. - Controls:
temperature,top_p,top_k,min_p,repetition_penalty,frequency_penalty,presence_penalty,stop,seed,response_format,logit_bias,logprobs,top_logprobs, andreasoning_effort, mapped to the nearest effort the model offers. - Output budget:
max_tokensormax_completion_tokens, lowered to the model's output limit; 16,384 when omitted. - Usage: set
stream_options.include_usageto receive token counts in the final chunk.X-Request-IDis the Ophiris usage ID shown in your billing history. - Not supported:
nother than 1, audio output, file inputs and remote image URLs are rejected. Any other field is dropped, so add-ons with separate charges never run unasked. - Web tools: the optional
web_toolsfield,{ "search": true, "readLinks": true, "youtube": false }, retrieves sources before generation. Each tool run is billed separately from the generation and reported as aophiris.toolevent. - On-chain analysis:
token_tools: truegives a tool-calling model 22 live analysis tools. 14 cover tokens on Solana, Ethereum, Base, BNB Chain and Robinhood Chain: search, overview, holders and security, holder groups, social activity, market-cap charts, on-chain contract reads, trending and hot boards, active tokens, signals, wallet profiles, market prices and chain activity. 8 cover tokenized stocks and real-world assets (index funds, treasuries, commodities, credit) on Robinhood Chain: search and session scans by asset class, quotes with the token's premium to its share and how unusual the move is, price charts, technical indicators, earnings with past reactions, news, and on-chain supply, holders, pools and the ERC-8056 share multiplier. Ophiris runs the model in rounds, executes its tools between them and returns one completion; your owntoolskeep working alongside. Token data is free; each model round is a separately billed request under the sameX-Request-ID. Lookups are reported asophiris.toolevents on streams and astool_runson JSON responses.
curl https://www.ophiris.xyz/api/v1/chat/completions \ -H "Authorization: Bearer $OPHIRIS_API_KEY" -H "Content-Type: application/json" \ -d '{"model": "google/gemma-4-26b-a4b-it-cybersecurity", "token_tools": true, "messages": [{"role": "user", "content": "What is trending on Solana right now, and is the top token safe to research?"}]}'Coding agents
Coding agents that accept an OpenAI-compatible provider work with Ophiris, including OpenCode, Cline, Roo Code, Kilo Code, Continue and Aider. Choose a model with tool calling. In the app, API access generates a ready opencode.json; the minimal form is:
{ "$schema": "https://opencode.ai/config.json", "provider": { "ophiris": { "npm": "@ai-sdk/openai-compatible", "name": "Ophiris", "options": { "baseURL": "https://www.ophiris.xyz/api/v1", "apiKey": "{env:OPHIRIS_API_KEY}" }, "models": { "google/gemma-4-26b-a4b-it-cybersecurity": { "name": "google/gemma-4-26b-a4b-it-cybersecurity", "tool_call": true } } } }}For Cline, Roo Code and Kilo Code, choose the OpenAI Compatible provider and enter the base URL, your key and a model ID.
MCP: Cursor and Claude Code
Ophiris is also an MCP server at https://www.ophiris.xyz/api/mcp (Streamable HTTP), so an agent can call Ophiris's tools directly while it codes. Nothing is installed locally.
- Free, no key:
list_modelsand the 22 on-chain tools: 14 for tokens, wallets and chains on Solana, Ethereum, Base, BNB Chain and Robinhood Chain, and 8 for tokenized stocks and real-world assets. They allow 60 calls an hour per address without a key and 60 a minute per key. - With a key:
asksends a prompt to one Ophiris model, for example a second opinion on code, andcomparesends it to two to four. They are billed like chat completions, under the key's limits, and report tokens, an estimated cost and the usage ID.balanceshows what is left on the key. - Fewer tools: add
?tools=models,token,stockor a comma-separated mix to the URL.
Cursor: install the Ophiris plugin (the server, research and code-review skills, and a rule), or add only the server. Set OPHIRIS_API_KEY in your environment before starting Cursor; without it the free tools still work. The server entry is:
{ "mcpServers": { "ophiris": { "url": "https://www.ophiris.xyz/api/mcp", "headers": { "Authorization": "Bearer ${env:OPHIRIS_API_KEY}" } } }}Claude Code:
claude mcp add --transport http ophiris https://www.ophiris.xyz/api/mcp \ --header "Authorization: Bearer $OPHIRIS_API_KEY"Other MCP clients use the same URL and header. A missing, wrong or expired key is reported in each tool result rather than as an HTTP error, so the agent can tell you what to fix.
Spending limits
- Per key: every key has a lifetime spending limit set when it is created. Requests that would pass it are refused with
key_limit_reached. A key can also be restricted to chosen models. - Per account: a monthly limit covers the workspace and every key together and is refused with
monthly_limit_reached. Change it in the app. - Holds: before a request starts, its largest possible cost (the prompt plus the full output budget at the model's rates) is held against the limits and the balance. The hold shrinks to the reported usage when the response ends and settles to the actual charge.
- Concurrency: each account has a limit on open API requests; further requests get
concurrency_limituntil one finishes.
Billing
Inference is charged per token at the rates shown for each model, in US dollars per million input and output tokens. Prompt-cache reads and writes use the model's cache rates where it publishes them. The app's Usage & billing lists every request with its model, tokens, charge and status.
On-chain balance deposits are not open yet.
Billing records stay private to your account. Itemized usage is never published as on-chain transactions.
Pay with OPHIRIS
Turn OPHIRIS into inference balance under Usage & billing in the app.
- Quote: choose an amount of balance. The quote states the exact OPHIRIS to send, the linked wallet to send it from and every cost below, and holds for 3 minutes.
- Payment: send the quoted OPHIRIS to the payment address
0x22023BAB2FBfb59cFc7aD52C32BEE7712aBA0Bb9. A payment that arrives after its quote expires, or for less than the quoted amount, is priced at the market when it is converted. - Conversion: Ophiris sells the OPHIRIS for ETH on the token's launchpad curve and credits the ETH it realized, in US dollars, when the sale settles. The credit is checked again once the payment's block is final.
- Fee: 1% of the payment is added on top. Half is burned by sending it to the dead address
0x…dEaD. The other half is kept by the payment address until OPHIRIS moves from its launchpad curve to its Uniswap v4 pool. After that it is added to the pool every hour as full-range liquidity whose position is owned by0x…dEaD, so it can never be withdrawn. - Other costs, shown in every quote: the curve's 1% launchpad fee and 1.95% creator tax (paid by the curve to the token's launch wallet), a 1% price buffer (what the sale does not use is credited back, up to 10% of the amount) and a 0.6% margin for moving the ETH to the inference provider.
- Limits: $1.00 to $100.00 per payment and $250.00 per account per day.
- Refunds: payments are converted when they arrive and are not refunded automatically. A payment that cannot be matched to an account or converted waits for review. A refund is sent by hand, and its transaction is recorded.
| Total | Value |
|---|---|
| Payments credited | 1 |
| Balance credited | $10.10 |
| OPHIRIS burned | 3,040 |
| OPHIRIS set aside for liquidity | 3,040 |
| OPHIRIS added to the pool as liquidity | 0 |
Recent burns: 3,040 OPHIRIS.
Totals, burns, liquidity adds and settlement batches, with their Robinhood Chain transactions: /api/token/payments.
Holder credits
Launch giveaway: the first 100 accounts whose wallet bought the Ophiris token in its trading pool, and still holds at least $1 of the tokens it bought, can claim $10.00 in inference credits under Holder credits in the app. Each account, wallet and purchase can claim once. Tokens received by transfer do not count.
Creator fees and daily AI credits
Since 16 Nov 2026 (transaction), OPHIRIS's creator fees from its launchpad are routed through the treasury 0x22023BAB2FBfb59cFc7aD52C32BEE7712aBA0Bb9, which claims them and splits each claim by the split in force:
- 50% to the team: sent in ETH to the deployer wallet
0x6E897AF4e3d280FBA9c5842BffdF1650F78FBb56. - 25% buyback and burn: the treasury buys OPHIRIS on its market and burns every token it bought by sending it to the dead address
0x…dEaD, which no one holds a key for. Burns before 17 Nov 2026 used the token's burn function, which also lowered total supply. - 25% AI credits for holders: moved to the inference provider and, once it arrives, shared every day, 00:00 UTC to the next 00:00 UTC, among the wallets that held OPHIRIS through that day, in proportion to each wallet's average balance over the day. A wallet needs at least 100,000 OPHIRIS on average. The launchpad's and the pool's contracts, the treasury, the team wallet and
0x…dEaDtake no share. Claim under Holder credits in the app with a linked wallet within 30 days; credit that is not claimed in time, and what rounding leaves, goes into the next day.
The split can change. Every split, the date it took effect and its reason, every round's transactions and every day's pool are listed in /api/token/payments; a change applies to claims after it, never to one already made.
Routing can also be turned off: the treasury can hand the creator fee recipient back to the deployer wallet, and to no other address. Every change of recipient, with its date and transaction, is listed in the same place.
| Creator fee claims | 18 |
| OPHIRIS bought back and burned | 13,020,245 |
| Set aside for holder AI credits | $179.00 |
Privacy
- Your workspace stays on your device. Conversations, projects, files and preferences are stored only in your browser, encrypted with AES-256-GCM. Each account has its own encrypted profile, which opens only after you sign in to that account, and its encryption key can never be exported from the browser.
- Sign-in details are not kept. Sign in with Google, X, GitHub or an Ethereum wallet. Ophiris shows your email, username or wallet back to you in the app but does not store the email or username; linked wallet addresses are kept only to match deposits, payments and holder credits to your account.
- No copies on our servers. Prompts and responses pass through Ophiris only to generate the answer; they are never stored.
- Billing keeps only what billing needs: the model, token counts, charge and status of each request, private to your account and never published on chain.
Network
Ophiris uses Robinhood Chain mainnet (chain ID 4663), an Ethereum layer 2 that pays network fees in ETH, for wallet sign-in, deposits and the Ophiris token.
Ophiris token contract: 0x78C1c2d4Cd6207B6FF4E84362d314D1cd0830041. This is the only official address; it is also linked from @OphirisLoop.
Errors
Errors use the OpenAI envelope, { "error": { "message", "type", "param", "code" } }. Errors during a stream arrive as a final data: frame with the same shape.
| Status | Code | Meaning |
|---|---|---|
| 400 | invalid_value, unsupported_parameter, unsupported_content, tools_not_supported | The request is malformed or asks for something the model or API does not support. |
| 400 | context_length_exceeded | The conversation is longer than the model's context window. |
| 401 | invalid_api_key | The key is missing, malformed or revoked. |
| 402 | insufficient_balance | The account balance cannot cover the request's hold. |
| 402 | monthly_limit_reached, key_limit_reached | The account's monthly limit or the key's spending limit would be exceeded. |
| 403 | model_not_allowed | The key is restricted to other models. |
| 404 | model_not_found | The model ID is not in the catalog. |
| 413 | request_too_large | The request body is over 4.4 MB. |
| 429 | concurrency_limit, upstream_rate_limited, rate_limit_exceeded | Too many open requests, a busy model, or the demo's per-visitor limit. Honor Retry-After. |
| 502 | upstream_error | The model provider failed. Retry or choose another model. |
| 503 | service_unavailable, demo_unavailable | Inference or the demo is temporarily unavailable. |