Models
GET https://api.devotel.com/v1/models{
"object": "list",
"data": [{ "id": "kimi-k3", "object": "model", "created": 0, "owned_by": "system" }]
}created is always 0 — the catalogue does not track model creation dates. It is present for
shape compatibility with clients that expect the field.
GET /v1/models lists what YOUR key can reach, not the whole catalogue. A key can be restricted
to specific models, and a few models are available only on request — so two keys can legitimately see
different lists. Pass any of the ids below as "model".
auto is in that list. Sending "model": "auto" lets the gateway pick the cheapest model
that can actually serve the request, keeps a conversation on the model it started with, and bills
you for the model that ran. See Auto (cost optimized).
The catalogue
| Model | ID | Type | Tools | Vision | In / 1M | Out / 1M |
|---|---|---|---|---|---|---|
| Auto (cost optimized) | auto | chat | Yes | Yes | — | — |
| Claude Haiku 4.5 | anthropic-claude-haiku-4.5 | chat | Yes | Yes | $1 | $5 |
| Claude Sonnet 5 | anthropic-claude-5-sonnet | chat | Yes | No | $2 | $10 |
| Claude Sonnet 5.5 | anthropic-claude-sonnet-5.5 | chat | Yes | Yes | $2 | $10 |
| Claude Sonnet 4.5 | anthropic-claude-4.5-sonnet | chat | Yes | Yes | $3 | $15 |
| Claude Sonnet 4.6 | anthropic-claude-4.6-sonnet | chat | Yes | Yes | $3 | $15 |
| Claude Opus 5.5 * | anthropic-claude-opus-5.5 | chat | Yes | Yes | $4 | $20 |
| Claude Opus 4.5 | anthropic-claude-opus-4.5 | chat | Yes | Yes | $5 | $25 |
| Claude Opus 4.6 | anthropic-claude-opus-4.6 | chat | Yes | Yes | $5 | $25 |
| Claude Opus 4.7 | anthropic-claude-opus-4.7 | chat | Yes | Yes | $5 | $25 |
| Claude Opus 4.8 | anthropic-claude-opus-4.8 | chat | Yes | Yes | $5 | $25 |
| Claude Opus 5 | anthropic-claude-opus-5 | chat | Yes | Yes | $5 | $25 |
| Claude Fable 5 * | anthropic-claude-fable-5 | chat | Yes | Yes | $10 | $50 |
| Claude Fable 5.1 * | anthropic-claude-fable-5.1 | chat | Yes | Yes | $10 | $50 |
| GPT-5 nano | openai-gpt-5-nano | chat | Yes | Yes | $0.05 | $0.4 |
| GPT-OSS 20B | openai-gpt-oss-20b | chat | Yes | No | $0.05 | $0.45 |
| GPT-OSS 120B | openai-gpt-oss-120b | chat | Yes | No | $0.06 | $0.39 |
| GPT-6 Luna | openai-gpt-6-luna | chat | Yes | Yes | $0.1 | $0.5 |
| GPT-4o mini | openai-gpt-4o-mini | chat | Yes | Yes | $0.15 | $0.6 |
| GPT-5.4 nano | openai-gpt-5.4-nano | chat | Yes | Yes | $0.2 | $1.25 |
| GPT-5.6 Luna | openai-gpt-5.6-luna | chat | Yes | Yes | $0.2 | $1.2 |
| GPT-5 mini | openai-gpt-5-mini | chat | Yes | Yes | $0.25 | $2 |
| GPT-5.4 mini | openai-gpt-5.4-mini | chat | Yes | Yes | $0.75 | $4.5 |
| o3-mini | openai-o3-mini | reasoning | Yes | No | $1.1 | $4.4 |
| GPT-5 | openai-gpt-5 | chat | Yes | Yes | $1.25 | $10 |
| GPT-5.2 | openai-gpt-5.2 | chat | Yes | Yes | $1.75 | $14 |
| GPT-5.3 Codex | openai-gpt-5.3-codex | coding | Yes | Yes | $1.75 | $14 |
| GPT-4.1 | openai-gpt-4.1 | chat | Yes | Yes | $2 | $8 |
| GPT-5.6 Terra | openai-gpt-5.6-terra | chat | Yes | Yes | $2 | $12 |
| GPT-6 Sol | openai-gpt-6-sol | chat | Yes | Yes | $2 | $10 |
| o3 | openai-o3 | reasoning | Yes | Yes | $2 | $8 |
| GPT-4o | openai-gpt-4o | chat | Yes | Yes | $2.5 | $10 |
| GPT-5.4 | openai-gpt-5.4 | chat | Yes | Yes | $2.5 | $15 |
| GPT-5.5 | openai-gpt-5.5 | chat | Yes | Yes | $5 | $30 |
| GPT-5.6 Sol | openai-gpt-5.6-sol | chat | Yes | Yes | $5 | $30 |
| GPT-6 Astra * | openai-gpt-6-astra | chat | Yes | Yes | $10 | $50 |
| o1 | openai-o1 | reasoning | Yes | Yes | $15 | $60 |
| GPT-5.2 Pro * | openai-gpt-5.2-pro | chat | Yes | Yes | $21 | $168 |
| GPT-5.4 Pro * | openai-gpt-5.4-pro | chat | Yes | Yes | $30 | $180 |
| DeepSeek 4 Flash | deepseek-4-flash | chat | Yes | No | $0.07 | $0.17 |
| DeepSeek V4 Flash (0731) | deepseek-v4-flash-0731 | chat | Yes | No | $0.08 | $0.25 |
| DeepSeek 3.2 | deepseek-3.2 | chat | Yes | No | $0.25 | $0.8 |
| DeepSeek V4 Pro (0813) | deepseek-v4-pro-0813 | chat | Yes | No | $1.32 | $3.96 |
| DeepSeek V4 Pro | deepseek-v4-pro | chat | Yes | No | $1.74 | $3.48 |
| Kimi K2.6 | kimi-k2.6 | chat | Yes | Yes | $0.95 | $4 |
| Kimi K3 (DigitalOcean) | kimi-k3 | reasoning | Yes | Yes | $3 | $15 |
| Qwen3.5 397B A17B | qwen3.5-397b-a17b | chat | Yes | Yes | $0.55 | $3.5 |
| Qwen3.8 Max | qwen3.8-max | chat | Yes | Yes | $2 | $6 |
| GLM-5.3 Flash | glm-5.3-flash | chat | Yes | Yes | $0.15 | $0.5 |
| GLM-5.2 | glm-5.2 | chat | Yes | No | $1.4 | $4.4 |
| GLM-5.3 | glm-5.3 | chat | Yes | No | $1.4 | $4.4 |
| Gemini 3.1 Flash-Lite | gemini-3.1-flash-lite | chat | Yes | Yes | $0.25 | $1.5 |
| Gemini 3.7 Flash | gemini-3.7-flash | chat | Yes | Yes | $0.75 | $3.75 |
| Gemini 3.8 Flash | gemini-3.8-flash | chat | Yes | Yes | $0.75 | $3.75 |
| Gemini 3.1 Pro | gemini-3.1-pro-preview | reasoning | Yes | Yes | $2 | $12 |
| Gemma 4 31B IT | gemma-4-31B-it | chat | Yes | Yes | $0.18 | $0.5 |
| Llama 4 Maverick | llama-4-maverick | chat | Yes | Yes | $0.2 | $0.7 |
| Mistral 3 14B | mistral-3-14B | chat | Yes | No | $0.2 | $0.2 |
| Nemotron Nano 12B v2 VL | nemotron-nano-12b-v2-vl | chat | Yes | No | $0.2 | $0.6 |
| MiniMax M2.5 | minimax-m2.5 | chat | Yes | No | $0.3 | $1.2 |
| MiMo v2.5 Pro | mimo-v2.5-pro | chat | Yes | No | $0.4 | $1.5 |
| Nemotron 3 Nano Omni | nemotron-3-nano-omni | chat | Yes | Yes | $0.5 | $0.9 |
| Nemotron 3 Ultra 550B | nemotron-3-ultra-550b | chat | Yes | No | $0.9 | $1.7 |
63 models. Prices are USD per million tokens; reasoning tokens are billed as output. Tools and vision are measured against the live provider, not taken from a datasheet. * Restricted: available on request — ask your administrator to grant it to a key.
Several of these reason before answering, even with no reasoning_effort field sent at all —
a small max_tokens can get cut off mid-thought and come back with no visible answer. See
Reasoning models and max_tokens for which
models, and the two fixes.
Model access is a per-key setting. By default a key may use every generally available model; you
or your administrator can narrow it to a specific list when the key is created, or later from the
admin panel. Asking for a model your key cannot use returns the same 404 model_not_found as
asking for one that does not exist — see Rate Limits & Errors.
kimi-k3
Moonshot AI’s Kimi K3, served through DigitalOcean Serverless Inference. It is the default and the most thoroughly audited model here, so the detail below is about it specifically; the table above covers the rest.
The same weights are also available directly from Moonshot, as kimi-k3-ms, added after
DigitalOcean’s kimi-k3 capacity proved unreliable during a service incident and DO was not able to
raise this account’s limits. It is a separate model id on purpose — a different provider means a
different prompt cache, a different invoice, and independent verification — so a session already
running on kimi-k3 is never silently moved there. auto will fail over to it automatically if
kimi-k3 hard-fails; you can also select kimi-k3-ms directly. reasoning_effort works exactly
as it does on kimi-k3, including none — measured live, not assumed. Moonshot bills a
cache-WRITE cost DigitalOcean does not; see Pricing below.
| Capability | Status |
|---|---|
| Tool calling / function calling | Yes |
Image input (image_url parts) | Yes |
Reasoning (reasoning_content) | Yes |
| Streaming (SSE) | Yes |
| Context window | 1,048,576 tokens |
Reasoning
Kimi K3 reasons before it answers. Every response carries reasoning_content next to content, and
those thinking tokens are billed as output tokens.
"message": {
"content": "Standing in between—\nI speak your words, not my own,\nfaceless messenger.",
"reasoning_content": "The user wants a haiku about proxies. Proxies could refer to..."
}Measured on this endpoint: that 72-character haiku cost 799 completion tokens, 782 of them
reasoning. A five-word greeting cost 158 completion tokens, 141 of them reasoning. Plan for
reasoning to dominate output cost on short answers, and read
Chat Completions before setting max_tokens.
Tool calling
Verified end to end: the model returns finish_reason: "tool_calls" with a tool_calls array,
accepts role: "tool" results, and then answers. Complete loops in Python and TypeScript are in
Tool Calling & Agents.
Image input
Send images as image_url content parts, either a data: URI or a URL the provider can fetch.
Verified against this endpoint with generated test images: the model named the shapes and their colours, read the embedded text, and counted objects correctly across three different images — for example “a red circle on the left and a blue triangle on the right, on a white background. Below them, in bold black text, it says: DEVOTEL 42”, and “three purple circles arranged in a row near the top, with a large black number 7 below them”.
{
"role": "user",
"content": [
{ "type": "text", "text": "What does this diagram show?" },
{ "type": "image_url", "image_url": { "url": "data:image/jpeg;base64,/9j/4AAQ..." } }
]
}A 900×640 image cost about 890 prompt tokens in these tests, so images are billed like a page of text rather than a rounding error.
Images count towards the request body, and /v1 bodies are capped at 20 MB — a base64 data
URL is about 4/3 the size of the file, so budget roughly 15 MB of actual image bytes across the
whole request, or pass a URL instead. Downscale before sending anyway: the model does not need
more than about 2000 px on the longest side, and every pixel is billed as prompt tokens.
Context window
The provider’s /v1/models metadata publishes no context length for kimi-k3 at all — unlike
kimi-k2.5/kimi-k2.6, which publish 262,144. Rather than leave K3 unstated indefinitely, we
measured it: a live probe on 2026-09-08 sent real requests at increasing sizes (max_tokens: 1,
filler content, cost tracked from the provider’s own prompt_tokens echo) until the provider
rejected one, then read the limit out of its own error text.
| Model | Context window | How we know |
|---|---|---|
kimi-k3 | 1,048,576 tokens | Measured 2026-09-08 — provider’s /v1/models publishes nothing for this one |
kimi-k2.5 / kimi-k2.6 | 262,144 tokens | Published by the provider’s /v1/models, confirmed by the same 2026-09-08 probe |
The number is the combined input + output budget, not input alone — the provider’s own message
below counted our one requested output token as part of the total. Exceeding it gets you exactly
this, verbatim, as a 400 invalid_request_error (the gateway passes a provider 400 straight
through rather than masking it as a 502 — see Rate Limits & Errors):
{
"error": {
"message": "This model's maximum context length is 1048576 tokens. However, you requested 1 output tokens and your prompt contains 1048708 input tokens, for a total of 1048709 tokens. Please reduce the length of the input prompt or the number of requested output tokens. (parameter=input_tokens, value=1048708)",
"type": "invalid_request_error",
"code": null,
"param": null
}
}No error observed at any size named a separate limit on output tokens alone — only the combined total above.
Pricing
Billed per token, at the rate for the direction of travel. Reasoning tokens are output tokens.
| Model | Input | Cache read | Output |
|---|---|---|---|
kimi-k3 | $3.00 / 1M tokens | $0.30 / 1M tokens | $15.00 / 1M tokens |
Cached input
When a request repeats a prompt prefix the provider has already seen, those tokens are served from its cache and billed at the cache read rate — an order of magnitude below the input rate. Two things follow:
usage.prompt_tokensincludes cached tokens; it is the total prompt size, not the billable part. Full-price input isprompt_tokens - cached_tokens.- The cached count arrives as
usage.prompt_tokens_details.cached_tokens.
Nothing needs enabling — caching is the provider’s, and a stable prefix (a long system prompt, a fixed set of tool definitions, a document you ask several questions about) is what earns the discount. Putting the variable part of a prompt last is what makes a prefix reusable.
The usage dashboard shows your daily token counts, the cached share, and an estimated cost computed from exactly these numbers.
kimi-k3-ms (Moonshot direct) has a cache-write cost DigitalOcean does not charge at all.
The first time a prompt prefix is seen it is billed as a cache WRITE rather than plain input — at
the default 5-minute TTL this happens to be the same rate as input, but at Moonshot’s 1-hour TTL
option it is double. The dashboard’s cost estimate for this model does not include the write
premium yet, so treat it as a lower bound rather than an exact figure until that formula is
proven against a real invoice.
Worked example
A request with 200 input tokens and 115 output tokens (reasoning included):
input: 200 / 1,000,000 × $3 = $0.00060
output: 115 / 1,000,000 × $15 = $0.00172
total ≈ $0.00232
Output is 5× the input rate, and reasoning lands on the output side — so on this model the cost of a request is driven by how hard it thinks, not by how long your prompt is.
Choosing a model
There is one. When more arrive they appear in GET /v1/models and on this page; existing model ids
keep working.