Skip to contentKumoDocs
Sections
On this page
Get started

Models

The platform publishes its own model catalog: what is available right now, how to name it in a request, what every capability means, and how to choose a model for your task.

View as Markdown

What the catalog is

The catalog is the list of models the platform can route right now. It is not written into this documentation and not maintained by hand: the page below reads it from the gateway, exactly as your code would.

curl https://api.kumorouter.com/v1/models

The answer is an OpenAI-shaped list: object: "list" and a data array whose every element is an object: "model". Reading it needs no key.

Not everything the platform holds gets in — only what can actually be called: the model is active and has at least one enabled capability in the published supplier configuration. A model's protocols and modalities are the exact intersection of what the catalog declares and what real supplier access has proven. So the list moves on its own when a supplier channel is withdrawn or added.

canonical_nameThe model's canonical name — what a request names it by.
aliasesThe aliases that lead to this same model.
vendorWho made the model — code, display names, brand color.
context_window_tokensHow many tokens the model accepts in one request. Absent when the catalog publishes no figure.
max_output_tokensHow many tokens it may answer with. Declared separately and never derived from the context window.
released_onThe day the model's maker released it, as `yyyy-mm-dd`. Not the day it was added here.
supplier_countDeprecated compatibility placeholder, always `1`. **Not** the number of suppliers behind a model — that is not published. Kept only so a client generated against an earlier release keeps parsing this listing; it will be removed.
protocol_codesThe protocols the model is reachable by.
modality_codesThe kind of work: text generation, embeddings, image generation.
capabilitiesWhat the model can do on each pair of protocol and modality — see below.
catalog_revisionA member of the list itself rather than of a model — the catalog revision, which grows on every change.
{
  "object": "model",
  "canonical_name": "<model>",
  "aliases": ["<alias>"],
  "vendor": { "code": "<vendor>" },
  "protocol_codes": ["chat_completions"],
  "modality_codes": ["text_generation"],
  "capabilities": [
    {
      "protocol_code": "chat_completions",
      "modality_code": "text_generation",
      "streaming_status": "proven",
      "tools_status": "unproven",
      "structured_output_mode": "json_schema"
    }
  ]
}

The catalog

Every model on the platform in one table, one row per model. The whole row opens that model's page; the single exception is the chip beside the identifier, which copies the name to the clipboard and leads nowhere.

ModelThe name, and under it the canonical name — exactly the string you will put in the `model` member. The maker stands beside it.
TypeWhat the model does: text, embeddings, images.
ProtocolsWhich protocols it can be called on.
ContextHow many tokens fit into one request together with its answer.
Max outputThe ceiling on the answer, in tokens.
InputThe price of a million input tokens.
OutputThe price of a million output tokens.
ReleasedThe day the platform dated the model's release.

The filters sit above the table. The search box looks at the name, the canonical name and the maker. Four rails follow, each narrowing on the thing it names: the maker, the type, the protocol, and a context window of "at least this much". The rails add up — choose a type and a protocol and you are asking for a model that does both. Pressing the chosen chip again clears it, and "Reset the filters" puts every rail back at once. The counter under them says how many rows are left, out of how many.

A column head sorts the table: the first press ascending, the second descending. A model the platform has published nothing for in that column sinks to the bottom in both directions — a dash means "not published", not "zero".

None of the figures in the table are written into this article. The platform serves the models and everything about them except money as a live catalog, and the rates as a published price list; the page reads both documents every time it is opened. So a model an operator adds or withdraws appears and disappears here on its own. If the price list cannot be read at this moment, the model list stays where it is and the rates stand as dashes.

Model catalog

A live list: the models come from the platform catalog and the rates from the published price list. A row opens the model’s own page; a column head sorts the table.

122 of 122 models
TypeProtocols
Claude Sonnet 5
claude-sonnet-5
Anthropic
TextMessages Chat Responses1M128K$1.2$6Jun 30, 2026
Gemini 3.6 Flash
gemini-3.6-flash
Google
TextChat1M65.5K$0.4$1.8Jul 21, 2026
Gemini 3.8 Flash
gemini-3.8-flash
Google
TextChat Responses1M65.5K$0.4$2.1Sep 2, 2026
GLM 5.3 Flash
glm-5.3-flash
Zhipu
TextChat Responses1M131.1K$0.2$0.6Aug 26, 2026
Kimi K2.8
kimi-k2-8
Moonshot AI
TextChat1M—$0.8$4Sep 11, 2026
Kimi K3
kimi-k3
Moonshot AI
TextChat Responses1M1M$1.7$8.3Jul 16, 2026
Kimi K3.1 Pro
kimi-k3-1-pro
Moonshot AI
TextChat1M943.7K$1.6$8Oct 23, 2026
Claude Opus 4.8
anthropic/claude-opus-4-8
Anthropic
TextMessages Chat Responses1M128K$2.6$12.8May 28, 2026
GPT Image 2
openai/gpt-image-2
OpenAI
ImagesImages————Apr 21, 2026
Claude Fable 5
anthropic/claude-fable-5
Anthropic
TextMessages Chat Responses1M128K$5.1$25.5Jun 7, 2026
Claude Fable 5.1
anthropic/claude-fable-5-1
Anthropic
TextMessages Chat Responses1M128K$6$30Sep 1, 2026
Claude Fable 5.2
anthropic/claude-fable-5-2
Anthropic
TextMessages Chat Responses1M128K$5.1$25.7Nov 9, 2026
Claude Haiku 4.5
anthropic/claude-haiku-4-5
Anthropic
TextMessages Chat Responses200K64K$0.6$3Oct 15, 2025
Claude Opus 4.5
anthropic/claude-opus-4-5-20251101
Anthropic
TextMessages Chat200K64K$2.6$12.8Nov 24, 2025
Claude Opus 4.6
anthropic/claude-opus-4-6
Anthropic
TextMessages Chat Responses1M128K$2.6$12.8Feb 4, 2026
Claude Opus 4.7
anthropic/claude-opus-4-7
Anthropic
TextMessages Chat Responses1M128K$2.6$12.8Apr 14, 2026
Claude Opus 5.5
anthropic/claude-opus-5-5
Anthropic
TextMessages Chat Responses1M128K$2.2$11Sep 22, 2026
Claude Opus 5
anthropic/claude-opus-5
Anthropic
TextMessages Chat Responses1M128K$2.6$12.8Jul 24, 2026
Claude Sonnet 5.5
anthropic/claude-sonnet-5-5
Anthropic
TextMessages Chat Responses1M128K$1.1$5.5Sep 28, 2026
Claude Sonnet 4.5
anthropic/claude-sonnet-4-5-20250929
Anthropic
TextMessages Chat200K64K$1.5$7.7Sep 29, 2025
Claude Sonnet 4.6
anthropic/claude-sonnet-4-6
Anthropic
TextMessages Chat Responses1M128K$1.5$7.7Feb 17, 2026
Gemini 3 Flash Preview
google/gemini-3-flash-preview
Google
TextChat Responses1M65.5K$0.3$1.5Dec 17, 2025
Gemini 3.1 Flash Lite
google/gemini-3.1-flash-lite
Google
TextChat1M65.5K$0.2$0.9May 7, 2026
Gemini 4 Carbon
google/gemini-4-carbon
Google
TextChat Responses1M32.8K$1.1$5.5Nov 10, 2026
Gemini 3.1 Pro Low
google/gemini-3.1-pro-low
Google
TextChat1M65.5K$1.1$6.6Feb 19, 2026
Gemini 4 Argon
google/gemini-4-argon
Google
TextChat Responses1M32.8K$1.1$5.5Oct 13, 2026
Gemini 3.6 Flash High
google/gemini-3.6-flash-high
Google
TextChat1M65.5K$0.4$1.9Jul 21, 2026
Gemini 3.7 Flash High
google/gemini-3.7-flash-high
Google
TextChat1M65.5K$0.4$1.9Aug 13, 2026
Gemini 3.8 Flash High
google/gemini-3.8-flash-high
Google
TextChat1M65.5K$0.5$2.3Sep 2, 2026
Codex Auto Review
openai/codex-auto-review
OpenAI
TextChat Responses1.1M128K$3$18—
GPT-5.3 Codex Spark
openai/gpt-5.3-codex-spark
OpenAI
TextChat128K32K$0.9$7.1Feb 5, 2026
GPT-5.5
openai/gpt-5.5
OpenAI
TextChat Responses1.1M128K$2.6$15.3Apr 23, 2026
GPT-5.6 Luna
openai/gpt-5.6-luna
OpenAI
TextChat Responses1.1M128K$0.1$0.6Jul 9, 2026
GPT-5.6 Sol
openai/gpt-5.6-sol
OpenAI
TextChat Responses1.1M128K$2$10.2Jul 9, 2026
GPT-6.1 Sol
openai/gpt-6.1-sol
OpenAI
TextChat Responses1.1M128K$1.1$5.5Sep 29, 2026
GPT-6.1 Terra
openai/gpt-6.1-terra
OpenAI
TextChat Responses1.1M128K$0.9$4.3Oct 26, 2026
GPT-6.2 Astra
openai/gpt-6.2-astra
OpenAI
TextChat Responses1.1M128K$4.8$24Nov 17, 2026
GPT-6.2 Sol
openai/gpt-6.2-sol
OpenAI
TextChat Responses1.1M128K$1.1$5.5Nov 17, 2026
GPT-6 Astra
openai/gpt-6-astra
OpenAI
TextChat Responses1.1M128K$6$30Sep 4, 2026
GPT-5.6 Terra
openai/gpt-5.6-terra
OpenAI
TextChat Responses1.1M128K$1$6.1Jul 9, 2026
GPT-6 Sol
openai/gpt-6-sol
OpenAI
TextChat Responses1.1M128K$1.1$5.5Sep 22, 2026
GPT-6 Luna
openai/gpt-6-luna
OpenAI
TextChat Responses1.1M128K$0.06$0.3Sep 22, 2026
Grok 4.5
xai/grok-4.5
xAI
TextChat Responses500K500K$1$3.1Jul 8, 2026
Grok 4.6
xai/grok-4.6
xAI
TextChat Responses500K500K$1$3.1Aug 12, 2026
Grok 4.7
xai/grok-4.7
xAI
TextChat Responses500K500K$1.1$3.3Sep 21, 2026
GLM-5.2
zai/glm-5.2
Zhipu
TextChat Responses1M131.1K$0.7$2.2Jun 13, 2026
GLM-5.3
zai/glm-5.3
Zhipu
TextChat Responses1M131.1K$0.8$2.6Aug 14, 2026
GLM-5.3 Raw
zai/glm-5.3-raw
Zhipu
TextChat1M131.1K$3$8.4Sep 14, 2026
Claude Haiku 5.5
anthropic/claude-haiku-5-5
Anthropic
TextMessages Chat1M128K$0.06$0.3Oct 7, 2026
Mistral Large 4
mistral-large-4
Mistral AI
TextChat1M262.1K$0.8$2.4Oct 6, 2026
Mistral Medium 3.5
mistral-medium-3-5
Mistral AI
TextChat262.1K209.7K$1.7$8.7Apr 30, 2026
Mistral Small 4
mistral-small-4
Mistral AI
TextChat262.1K209.7K$0.2$0.8Mar 16, 2026
Devstral 2
devstral-2
Mistral AI
TextChat262.1K209.7K$0.5$2.3Dec 9, 2025
DeepSeek V4.1 Flash
deepseek-v4-1-flash
DeepSeek
TextChat1M393.2K$0.1$0.5Sep 10, 2026
DeepSeek V4 Pro 0813
deepseek-v4-pro-0813
DeepSeek
TextChat1M393.2K$0.5$1.6Aug 12, 2026
DeepSeek V4 Flash
deepseek-v4-flash
DeepSeek
TextChat1M943.7K$0.07$0.1Apr 24, 2026
Qwen3.8 Max 0902
qwen3-8-max-0902
Qwen
TextChat1M131.1K$2.3$7Sep 3, 2026
Qwen3.8 Max Preview
qwen3-8-max-preview
Qwen
TextChat1M131.1K$2.3$7Jul 19, 2026
Qwen3.7 Max
qwen3-7-max
Qwen
TextChat1M131.1K$1.7$5.1May 21, 2026
Qwen3.8 Flash
qwen3-8-flash
Qwen
TextChat1M131.1K$0.2$0.5Aug 26, 2026
Qwen3.8 Omni Flash
qwen3-8-omni-flash
Qwen
TextChat1M131.1K$0.2$0.5Sep 21, 2026
Qwen Image 2.1 Pro
qwen-image-2-1-pro
Qwen
ImagesImages————Oct 17, 2026
Qwen Image 2.1 Turbo
qwen-image-2-1-turbo
Qwen
ImagesImages————Oct 17, 2026
Qwen Image 3
qwen-image-3
Qwen
ImagesImages————Aug 5, 2026
Qwen Image 3 Pro
qwen-image-3-pro
Qwen
ImagesImages————Aug 5, 2026
Qwen3.8 27B
qwen3-8-27b
Qwen
TextChat1M235.9K$0.05$2.9Aug 14, 2026
Qwen3.8 27B Antislop
qwen3-8-27b-antislop
Qwen
TextChat262.1K16.4K$0.7$6Oct 3, 2026
Qwen3.7 Plus
qwen3-7-plus
Qwen
TextChat1M131.1K$0.4$1.5Jun 3, 2026
Qwen3 Coder Next
qwen3-coder-next
Qwen
TextChat262.1K235.9K$0.1$0.9Feb 4, 2026
MiniMax M3
minimax-m3
MiniMax
TextChat1M512K$0.3$1.4May 31, 2026
MiniMax M2.7
minimax-m2-7
MiniMax
TextChat204.8K176.9K$0.3$1.4Mar 18, 2026
Kimi K2.7 Code
kimi-k2-7-code
Moonshot AI
TextChat262.1K235.9K$0.8$3.9Jun 12, 2026
Kimi K2.6
kimi-k2-6
Moonshot AI
TextChat262.1K235.9K$0.8$4Apr 20, 2026
GLM-5.3 FlashX
glm-5-3-flashx
Zhipu
TextChat1M131.1K$0.4$1.5Sep 18, 2026
MiMo V2.6 Pro
mimo-v2-6-pro
Xiaomi
TextChat1.1M131.1K$0.5$1Sep 21, 2026
MiMo V2.6 Flash
mimo-v2-6-flash
Xiaomi
TextChat1.1M131.1K$0.2$0.3Sep 21, 2026
Step 5 Preview
step-5-preview
StepFun
TextChat1M64K$1.2$3.1Oct 8, 2026
Step 3.7 Flash
step-3-7-flash
StepFun
TextChat262.1K230.4K$0.2$1.3May 28, 2026
Muse Glimmer 30B
muse-glimmer-30b
Meta
TextChat131.1K16.4K$0.3$1.4Aug 9, 2026
Nemotron 3 Ultra
nemotron-3-ultra
NVIDIA
TextChat262.1K182.5K$0.7$2.8Jun 4, 2026
Nemotron 3 Super
nemotron-3-super
NVIDIA
TextChat262.1K235.9K$0.09$0.5Mar 11, 2026
Seed 2.1 Turbo
seed-2-1-turbo
ByteDance Seed
TextChat262.1K235.9K$0.6$2.9Aug 12, 2026
Seed 2.0 Code
seed-2-0-code
ByteDance Seed
TextChat262.1K131.1K$0.6$3.5Aug 12, 2026
Hy4 Preview
hy4-preview
Tencent
TextChat1M64K$1$2.9Aug 28, 2026
Command A+
command-a-plus
Cohere
TextChat192K64K$0.3$1.7Sep 22, 2026
Inkling
inkling
Thinking Machines
TextChat524.3K471.9K$1.2$4.7Jul 17, 2026
LongCat 2.0
longcat-2-0
Meituan
TextChat1M262.1K$0.3$1.4Jul 20, 2026
Nova 2 Lite
nova-2-lite
Amazon
TextChat1M65.5K$0.3$2.9Dec 2, 2025
Solar Pro 4
solar-pro-4
Upstage
TextChat524.3K131.1K$0.1$0.4Aug 10, 2026
Aion 3.5
aion-3-5
AionLabs
TextChat262.1K32.8K$3.5$7Sep 23, 2026
Mercury 2.5
mercury-2-5
Inception
TextChat260K65.5K$0.05$0.2Sep 8, 2026
Llama 4 Maverick
llama-4-maverick
Meta
TextChat1M16.4K$0.2$0.8Apr 5, 2025
Granite 4.2 8B
granite-4-2-8b
IBM
TextChat131.1K118K$0.07$0.3Aug 31, 2026
GPT Image 2.5
openai/gpt-image-2.5
OpenAI
ImagesImages—————
Clef
clef
Cloudflare
DecisionsSystem One65.5K—$0.05—Oct 1, 2026
Clef Flash
clef-flash
Cloudflare
DecisionsSystem One24.6K—$0.05—Oct 1, 2026
Clef Omni
clef-omni
Cloudflare
DecisionsSystem One65.5K—$0.2—Oct 9, 2026
D1
liquid-d1
Liquid AI
DecisionsSystem One65.5K—$0.02—Oct 1, 2026
D1 Ultra
liquid-d1-ultra
Liquid AI
DecisionsSystem One32.8K—$0.05—Oct 15, 2026
Jev 1.13
jev-1-13
TypeSafe
DecisionsSystem One262.1K—$0.05—Sep 18, 2026
Jev 1.14
jev-1-14
TypeSafe
DecisionsSystem One262.1K—$0.05—Oct 14, 2026
Microsoft-Decision-1
microsoft-decision-1
Microsoft
DecisionsSystem One32.8K—$0.05—Oct 9, 2026
Solar Decide
solar-decide
Upstage
DecisionsSystem One524.3K—$0.06—Sep 28, 2026
Solar Decide Flash
solar-decide-flash
Upstage
DecisionsSystem One524.3K—$0.06—Oct 8, 2026
Span-01
span-01
Respan
DecisionsSystem One——$0.02—Sep 26, 2026
Span-02
span-02
Respan
DecisionsSystem One131.1K—$0.01—Oct 16, 2026
Decider V1.2
pplx-decider-v1-2
Perplexity
DecisionsSystem One262.1K—$0.02—Oct 16, 2026
Decider Flash V1.2
pplx-decider-flash-v1-2
Perplexity
DecisionsSystem One262.1K—$0.01—Oct 16, 2026
Drex v1.5
drex-v1-5
Nace.AI
DecisionsSystem One131.1K—$0.05—Oct 9, 2026
Drex v2.0
drex-v2-0
Nace.AI
DecisionsSystem One131.1K—$0.07—Oct 12, 2026
Classify-1
classify-1
ScaleDown
DecisionsSystem One1M—$0.05—Oct 13, 2026
Mercury Decide
mercury-decide
Inception
DecisionsSystem One65.5K—$0.02—Sep 30, 2026
Tev1 4B Experimental
tev1-4b-experimental
Together AI
DecisionsSystem One32.8K—$0.05—Sep 30, 2026
Kev 4B
kev-4b
Jared Palmer
DecisionsSystem One8.2K—$0.05—Sep 25, 2026
GPT Image 1.5
openai/gpt-image-1.5
OpenAI
ImagesImages————Nov 25, 2025
GPT Image 1
openai/gpt-image-1
OpenAI
ImagesImages————Apr 24, 2025
GPT Image 2.5 Flare
openai/gpt-image-2.5-flare
OpenAI
ImagesImages—————
GPT Image 2.5 Sunburst
openai/gpt-image-2.5-sunburst
OpenAI
ImagesImages—————
Grok Imagine Lite
xai/grok-imagine-image-lite
xAI
ImagesImages—————
OpenJev
openjev
RawModels
DecisionsSystem One32.8K—$0.05——
OpenJev Small
openjev-small
RawModels
DecisionsSystem One32.8K—$0.03——
GPT Image 2.6 Sunburst
openai/gpt-image-2.6-sunburst
OpenAI
ImagesImages—————
Every model rate

How to name a model

A request names the model as a string in the model member. Either the canonical name or any alias that leads to the same model will do.

  • A canonical name belongs to one model and to no other.
  • An alias is a second name for the same model. At any moment an alias leads to exactly one model; over time it can be pointed elsewhere, and that is the whole point of it — code that names an alias moves with it.
  • Canonical names and aliases live in one namespace: the same string cannot be both, and cannot belong to two models.

A name no live model answers to is refused straight away — the platform says plainly that it has no such model instead of guessing at one.

Note

A key may carry a scope. If that scope lists models, the key may name only those: a scope is a whitelist, not a preference. A call that names something outside the list is refused rather than routed.

What the platform records about a key →

How to read the capabilities

Capabilities are not declared for a model in general. They are declared for each pair of a protocol and a modality, so one model can do more on one protocol than on another.

Protocols

chat_completionsThe OpenAI-shaped conversational operation: messages in, `choices` out.
responsesThe second operation of that ecosystem, with a request and answer shape of its own.
anthropic_messagesThe Anthropic-native shape, understood by clients and SDKs written against that API.
embeddingsVectors for text.
imagesImage generation.

Modalities

A modality is the kind of work rather than the shape of the request: text_generation, embeddings, image_generation. The protocol says how you ask; the modality says what you get. A key's scope restricts modalities, which is why the pair of them decides whether a call is allowed at all.

The three states

Every capability — streaming, tools, structured output, strict semantics, streaming of structured output — stands in one of three states.

StateWhat it meansWhat the gateway does
ProvenThe platform checked it on a real supplier channel.The call goes upstream.
Not provenThe platform has not checked it. Not "it fails", but "we have not proven it".A request that needs the capability is refused before the supplier is contacted.
UnsupportedThe capability is not there.Such a request is refused in the same way.

The difference between "not proven" and "unsupported" is in where it comes from, not in what you see: both are refused alike and refused early. The refusal happens without reaching the supplier and does not change if you repeat the same request — if you need a capability, pick a model that has it proven.

Structured output carries a mode as well: json_object (the answer must be a JSON object) or json_schema (the answer must match your schema); none means there is no mode at all.

How a response schema is stated → Tools →

The model page

Every model in the catalog has a page of its own at /models/<canonical name>. It collects what you would otherwise assemble from several sources:

  • the figures: context window, output ceiling and release date;
  • a quickstart with this model's identifier already substituted — copy the code as it stands;
  • a capability matrix per protocol: what is proven and what is not;
  • the wallet's prices per dimension;
  • links onward to the integration pages with that model chosen.

The catalogue table above leads to these pages, and the documentation search indexes them.

How to choose a model

The catalog answers "what is available"; the choosing is yours. An order that saves time:

  1. Filter by protocolKeep the models that speak the protocol your client speaks. Changing the client costs more than changing the model.
  2. Check the capabilitiesIf you need tools, streaming or a strict response schema, the capability has to be proven rather than merely mentioned.
  3. Compare the volumesThe context window decides how much you can send, the output ceiling how much you can get back. They are two independent figures.
  4. Work out the priceRates are published per million tokens, separately for input and for output, so no single number compares two models.
  5. Check measured performanceThe model card shows request latency, generation speed and availability over the last 24 hours. Missing measurements are not a zero delay or perfect availability.
  6. Note the release dateIt is the day the maker released the model, so it tells you the model's generation, not when it was added here.

Do not choose based on a single request. Take a dozen of your real tasks, run two or three candidates over them, and compare the answers and the spend — the catalog gives you a shortlist, your own run decides.

The first call → Limits →