Smoo AI LLM Gateway

79+ frontier models, one endpoint.

Drop your OpenAI SDK's base URL to https://llm.smoo.ai/v1. Same shapes, same streaming, same tool calls — with unified billing, org-scoped keys, and cross-lab fallback chains that catch 429s, 5xxs, and timeouts before they hit your code.

15 frontier42 smart12 fastOpenAI · Anthropic · Google · Groq · DeepSeek · Qwen · Kimi · GLM · MiniMax

Get an API key Browse models

Drop-in quickstart

Same OpenAI SDK you already use. Different base URL and your Smoo AI virtual key. That's it.

curl

curl https://llm.smoo.ai/v1/chat/completions \
  -H "Authorization: Bearer $SMOOAI_LLM_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gemini-2.5-flash",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

python

from openai import OpenAI

client = OpenAI(
    api_key=os.environ["SMOOAI_LLM_KEY"],
    base_url="https://llm.smoo.ai/v1",
)

resp = client.chat.completions.create(
    model="gemini-2.5-flash",
    messages=[{"role": "user", "content": "Hello"}],
)
print(resp.choices[0].message.content)

typescript

import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: process.env.SMOOAI_LLM_KEY,
  baseURL: 'https://llm.smoo.ai/v1',
});

const resp = await client.chat.completions.create({
  model: 'gemini-2.5-flash',
  messages: [{ role: 'user', content: 'Hello' }],
});
console.log(resp.choices[0].message.content);

Model catalog

Pricing shown is passthrough cost in USD per million tokens, used as the basis for org-metered overage. See /pricing for plan-included allowances and volume tiers.

Model	Family	Tier	Context	Input / 1M	Output / 1M	Best for
claude-opus-4-6	Anthropic	Frontier	200K	$15.00	$75.00	Deepest multi-step reasoning, long-horizon planning, high-fidelity code
claude-sonnet-4-6	Anthropic	Smart	200K	$3.00	$15.00	Best tool-use + diff fidelity in our coding tests (BFCL v3, τ²-bench)
claude-sonnet-4-5	Anthropic	Smart	200K	$3.00	$15.00	Sonnet 4.5 — kept available for prompts pinned before 4.6
claude-haiku-4-5	Anthropic	Fast	200K	$1.00	$5.00	Cheap, fast, strong JSON adherence — good for judges + classifiers
claude-opus-4-7	Anthropic	Frontier	200K	$5.00	$25.00	64.3% SWE-bench Pro — step-change agentic coding over 4.6 (GA 2026-04-16)
claude-opus-4-8	Anthropic	Frontier	1M	$5.00	$25.00	Current Anthropic flagship — 1M context, ~4× less likely than 4.7 to slip a flaw (GA 2026-05-28) Fast mode available at $10/$25 per 1M for ~2.5× speed
gpt-5.5	OpenAI	Frontier	400K	$5.00	$30.00	Current OpenAI flagship — GA 2026-04-24, reduced hallucination on regulated domains
gpt-5.5-pro	OpenAI	Frontier	400K	$30.00	$180.00	GPT-5.5 Pro reasoning tier — most capable, highest cost
gpt-5.4	OpenAI	Frontier	400K	$2.50	$15.00	Mid-frontier GPT-5.4 — between gpt-5 and gpt-5.5 on capability + cost
gpt-5.4-mini	OpenAI	Smart	400K	$0.75	$4.50	Cheaper smart-tier GPT-5.4 sibling
gpt-5.4-nano	OpenAI	Fast	400K	$0.20	$1.25	Ultra-cheap GPT-5.4-nano — high-volume structured output
gpt-5.4-pro	OpenAI	Frontier	400K	$30.00	$180.00	GPT-5.4 Pro reasoning tier
gpt-5.2	OpenAI	Smart	400K	$1.75	$14.00	GPT-5.2 — mid-tier between 5.1 and 5.4
gpt-5.2-codex	OpenAI	Smart	400K	$1.75	$14.00	GPT-5.2 with Codex-coding training
gpt-5.2-pro	OpenAI	Frontier	400K	$21.00	$168.00	GPT-5.2 Pro reasoning tier
gpt-5.1	OpenAI	Smart	400K	$1.25	$10.00	GPT-5.1 — refined GPT-5 family entry
gpt-5.1-codex	OpenAI	Smart	400K	$1.25	$10.00	GPT-5.1 with Codex-coding training
gpt-5	OpenAI	Frontier	256K	$2.50	$10.00	Prior flagship — kept for pinned prompts
gpt-5-pro	OpenAI	Frontier	256K	$15.00	$120.00	GPT-5 Pro reasoning tier
gpt-5-codex	OpenAI	Smart	256K	$1.25	$10.00	GPT-5 with Codex-coding training
gpt-5-mini	OpenAI	Smart	256K	$0.50	$2.00	Balanced smart-tier option with GPT-5 training
gpt-5-nano	OpenAI	Fast	400K	$0.20	$1.25	Cheapest GPT-5 variant, good for high-volume structured output
gpt-4.1	OpenAI	Smart	1M	$2.00	$8.00	Big context, strong coding + tool use
gpt-4.1-mini	OpenAI	Fast	1M	$0.40	$1.60	Long-context, low-cost workhorse
gpt-4.1-nano	OpenAI	Fast	1M	$0.10	$0.40	Ultra-cheap 1M-context option for ingestion + summaries
gpt-4o	OpenAI	Smart	128K	$2.50	$10.00	Mature multimodal (text + image); good for stable prompts
gpt-4o-mini	OpenAI	Fast	128K	$0.15	$0.60	Battle-tested cheap tier — wide SDK compatibility
o3	OpenAI	Specialty	200K	$2.00	$8.00	o-series reasoning — visible chain-of-thought, strong on math + logic
o3-mini	OpenAI	Specialty	200K	$1.10	$4.40	Cheaper o-series reasoning sibling
o3-pro	OpenAI	Specialty	200K	$20.00	$80.00	o3 Pro — extended reasoning for hardest problems
o4-mini	OpenAI	Specialty	200K	$1.10	$4.40	Reasoning-optimized — strong at math, logic, code synthesis
omni-moderation-latest	OpenAI	Specialty	32K	Free	Free	Free content safety classifier — used by built-in guardrails Free from OpenAI; Smoo passes through at cost
gemini-3.5-flash	Google	Smart	1M	$1.50	$9.00	Current Google flagship Flash — GA 2026-05-19, 76.2% Terminal-Bench 2.1
gemini-2.5-pro	Google	Frontier	1M	$1.25	$10.00	Frontier reasoning with 1M context; great for large-doc analysis
gemini-2.5-flash	Google	Smart	1M	$0.30	$2.50	Best tool-use-per-dollar (BFCL v3 leader in its price band) Smoo AI default smart model
gemini-2.5-flash-lite	Google	Fast	1M	$0.10	$0.40	Very cheap, 1M context, fast first-token
gemini-2.0-flash	Google	Fast	1M	$0.10	$0.40	Stable 2.0 family — kept for pinned prompts
gemini-3-flash-preview	Google	Smart	1M	—	—	Next-gen Flash preview — 3/3 PASS on CS escalation E2E Preview pricing not yet published
gemini-3.1-flash-lite	Google	Fast	1M	—	—	GA 3.1 Flash-Lite — 2.1s TTFT latency champion, voice-pipeline-tier Pricing not yet posted in our catalog; check the dashboard for live rate
gemini-3.1-flash-lite-preview	Google	Fast	1M	—	—	Preview alias retained for backwards-compat with pinned callers Use gemini-3.1-flash-lite (GA) for new code
gemini-3-pro-preview	Google	Frontier	1M	—	—	Next-gen Pro preview Preview pricing not yet published
gemini-3.1-pro-preview	Google	Frontier	1M	—	—	Next-gen Pro refresh preview Preview pricing not yet published
groq-llama-3.3-70b	Groq	Smart	128K	$0.59	$0.79	Llama 3.3 70B on Groq — fast, cheap, clean tool loops
groq-llama-3.1-8b	Groq	Fast	128K	$0.050	$0.080	Sub-300ms first token; cheapest path through Groq Smoo AI default fast model (used for voice pipeline)
groq-llama-4-scout	Groq	Smart	10M	$0.11	$0.34	10M context — ingestion + large document reasoning
groq-llama-4-maverick	Groq	Smart	1M	$0.20	$0.60	Larger Llama 4 variant, stronger reasoning than Scout
groq-kimi-k2	Groq	Smart	128K	$1.00	$3.00	Kimi K2-Instruct — MoE design, strong agentic task quality
groq-gpt-oss-120b	Groq	Smart	128K	$0.15	$0.60	OpenAI OSS 120B — best for single-turn generation Not recommended for multi-turn tool loops; known to drop structured output
groq-gpt-oss-20b	Groq	Fast	128K	$0.10	$0.30	OpenAI OSS 20B — cheap, fast single-shot generation Not recommended for multi-turn tool loops
groq-gpt-oss-safeguard-20b	Groq	Specialty	128K	$0.10	$0.30	Safety-tuned open-weight GPT — content moderation tasks
deepseek-v4-flash	DeepSeek	Smart	1M	$0.14	$0.28	1M context, dual Thinking/Non-Thinking modes — smooth-reasoning primary
deepseek-v4-pro	DeepSeek	Smart	1M	$0.43	$0.87	Pro-tier V4 reasoner — 75% intro discount through 2026-05-31 List price $1.74/$3.48 per 1M; refresh when intro ends
deepseek-chat	DeepSeek	Smart	1M	$0.14	$0.28	Legacy alias — routes to deepseek-v4-flash (retiring 2026-07-24)
deepseek-reasoner	DeepSeek	Smart	1M	$0.43	$0.87	Legacy alias — routes to deepseek-v4-pro (retiring 2026-07-24)
qwen-3.7-max-direct	Alibaba DashScope	Frontier	1M	$2.50	$7.50	Current Qwen flagship — agent-first, native thinking, 200 tok/s, SWE-Pro + Terminal-Bench tier winner (GA 2026-05-20) 90% cache-hit discount ($0.25/M); accepts both OpenAI ChatCompletions and Anthropic Messages format
qwen-3.6-plus-direct	Alibaba DashScope	Smart	1M	$0.33	$1.95	Qwen 3.6 Plus generalist — GA 2026-04-02, 1M context
qwen3-coder-flash-direct	Alibaba DashScope	Smart	1M	$0.30	$1.50	Bench-winning coder — smooth-coding primary, 16/16 aider-polyglot PASS
qwen3-coder-plus-direct	Alibaba DashScope	Smart	1M	$1.00	$5.00	PR-review tuned coder for large diffs
kimi-k2.6-direct	Moonshot	Smart	262K	$0.95	$4.00	Current Kimi flagship — ties GPT-5.5 on SWE-Bench Pro, GA 2026-04-20
kimi-k2-thinking-direct	Moonshot	Smart	256K	$0.60	$2.50	Deepest reasoner in the Kimi line — smooth-reasoning fallback
kimi-k2.5-direct	Moonshot	Smart	256K	$0.60	$2.50	Flagship general-purpose Kimi via Moonshot direct
glm-5.1-direct	Z.ai	Smart	200K	$0.60	$2.20	58.4% SWE-Pro — coder-forward, smooth-coding fallback (GA 2026-04-07)
glm-5-direct	Z.ai	Smart	200K	$0.60	$1.92	Faster GLM (78 tok/s vs 5.1's 54) — GA 2026-02-11
minimax-m2-direct	MiniMax	Smart	200K	$0.30	$1.20	Frontier-class reviewer at $0.30 input — smooth-reviewing primary
minimax-m2.7-direct	MiniMax	Smart	200K	$0.30	$1.20	Current MiniMax flagship — same price as M2
minimax-m2.7-highspeed-direct	MiniMax	Smart	200K	$0.60	$2.40	Throughput-optimized M2.7 — ~100 tps
deepseek-v3.2	DeepSeek (via aggregator)	Smart	128K	$0.27	$1.10	Aggregator-routed V3.2 — emergency failover only
deepseek-r1	DeepSeek (via aggregator)	Frontier	64K	$0.55	$2.19	Aggregator-routed R1 reasoner — emergency failover only
glm-5.1	Z.ai (via aggregator)	Smart	128K	$0.60	$2.20	Aggregator-routed GLM 5.1 — emergency failover only
minimax-m2.7	MiniMax (via aggregator)	Smart	200K	$0.30	$1.20	Aggregator-routed M2.7 — emergency failover only
minimax-m2.5	MiniMax (via aggregator)	Smart	200K	$0.30	$1.20	Aggregator-routed M2.5 — emergency failover only
kimi-k2.5	Moonshot (via aggregator)	Smart	256K	$0.60	$2.50	Aggregator-routed K2.5 — emergency failover only
kimi-k2-thinking	Moonshot (via aggregator)	Smart	256K	$0.60	$2.50	Aggregator-routed K2-Thinking — emergency failover only
qwen3-coder-plus	Alibaba (via aggregator)	Smart	1M	$1.00	$5.00	Aggregator-routed Coder-Plus — emergency failover only
qwen3-coder-flash	Alibaba (via aggregator)	Smart	1M	$0.30	$1.50	Aggregator-routed Coder-Flash — emergency failover only
text-embedding-3-small	OpenAI	Embedding	8K	$0.020	—	1536-dim embeddings — Smoo AI default for knowledge base ingestion
text-embedding-3-large	OpenAI	Embedding	8K	$0.13	—	3072-dim embeddings — higher retrieval quality for specialist corpora
gemini-embedding-001	Google	Embedding	8K	$0.15	—	3072-dim Gemini embeddings — strong on multilingual + code
gemini-embedding-002	Google	Embedding	8K	—	—	First natively multimodal embedding — text + images + video + audio in one space, Matryoshka 3072→1536→768 dims (GA 2026-04-23) Strict upgrade over -001 for new RAG surfaces; keep -001 for index compatibility

Prices refresh as upstream labs publish changes — your dashboard shows live rates and the effective rate after your plan's tier allowance. Overage is billed per your subscription tier.

Smooth semantic aliases

Point the Smooth coding runtime at llm.smoo.ai/v1 and use stable intent-based model names. Aliases re-target to new upstream models as better options ship — your code doesn't change.

Alias	Resolves to	Purpose
smooth-coding	qwen3-coder-flash	Coding workhorse — best agentic tool-use, native OpenAI tool_calls, 1M ctx
smooth-reasoning	deepseek-v4-flash	Deep reasoning + planning — 1M ctx, dual Thinking/Non-Thinking, $0.14/$0.28
smooth-reviewing	MiniMax-M2	Adversarial critique — different lab from the coder
smooth-judge	groq/llama-3.1-8b-instant	Sub-300ms JSON judge for guardrails + Narc verdicts
smooth-summarize	gemini-2.5-flash	Long-context summaries + compression (1M ctx)
smooth-planning	gemini-2.5-flash	Structured mapper / planning flows
smooth-fast	groq/llama-3.1-8b-instant	Sub-300ms utility — session naming, titles, autocomplete, voice fast-router
smooth-thinking	deepseek-v4-flash	Deprecated alias for smooth-reasoning (legacy callers)

Every alias has a fallback chain — a single provider outage degrades to the next-best option rather than failing the request.

Smooth Routing — bench-backed primaries, cross-vendor fallback chains

Each smooth-* alias points at the current bench-leader for that purpose and falls back through a hand-curated chain that spans at least three labs. A Vertex 503 or DeepSeek hiccup degrades to the next provider before the caller sees an error. Re-evaluated each time the published-benchmark picture shifts (most recent refresh: May 2026).

smooth-codingCoding workhorse — code generation, diffs, agentic tool loops.

Primary

qwen3-coder-flash

DashScope (Alibaba)

Why this primary

Best agentic tool-use in the cheap-coder tier — native OpenAI tool_calls with no thinking-mode contract. DeepSeek-V4 leads SWE-bench Verified at 80.6% but its thinking-mode trips LiteLLM's reasoning_content guard on turn 1, so we pin Qwen as the entry-point primary and let DeepSeek-class reasoning land in smooth-reasoning. 1M context, $0.30/$1.50.

Fallback chain (in order)

glm-5.1 (Z.ai)→
kimi-k2-thinking (Moonshot)→
MiniMax-M2 (MiniMax)

smooth-reasoningDeep multi-step reasoning + planning.

Primary

deepseek-v4-flash

DeepSeek

Why this primary

1M context, dual Thinking / Non-Thinking modes, $0.14/$0.28 per 1M — the cheapest top-tier reasoner.

Fallback chain (in order)

kimi-k2-thinking (Moonshot)→
qwen3-235b-thinking-2507 (Alibaba)

smooth-reviewingAdversarial critique of generated code (PR review, bug-spotting).

Primary

MiniMax-M2

MiniMax

Why this primary

Cheap ($0.30/$1.20), coding-forward, agentic. Different lab than the coder — catches bugs the coder's training didn't see.

Fallback chain (in order)

qwen3-coder-plus (Alibaba)→
glm-5.1 (Z.ai)→
kimi-k2-thinking (Moonshot)

smooth-judgePer-tool JSON verdicts (guardrails, Narc, content gates).

Primary

groq/llama-3.1-8b-instant

Groq

Why this primary

Sub-300ms first-token. Verdicts are 1-line JSON; Gemini Flash was overkill at ~10× the cost. Matches accuracy in the small-verdict shape.

Fallback chain (in order)

gemini-2.5-flash (Google)→
claude-haiku-4-5 (Anthropic)→
gpt-5-mini (OpenAI)

smooth-summarizeTranscript + long-document compression.

Primary

gemini-2.5-flash

Google

Why this primary

1M context, IFEval leader in the cheap Gemini tier, dialable thinking levels.

Fallback chain (in order)

qwen3-coder-plus (Alibaba, 1M ctx backstop)→
gpt-5-mini (OpenAI)

smooth-fastLatency-critical utility — session titles, autocomplete, voice fast-router.

Primary

groq/llama-3.1-8b-instant

Groq

Why this primary

Sub-300ms first-token; P95 ~600ms end-to-end on production traffic. ~10× cheaper than the prior Gemini Flash Lite primary, ~1s faster on cold starts.

Fallback chain (in order)

gemini-2.5-flash-lite (Google)→
claude-haiku-4-5 (Anthropic)→
gpt-5-mini (OpenAI)

smooth-planningStructured-plan mapper — read-only planning flows.

Primary

gemini-2.5-flash

Google

Why this primary

Long context, structured-output friendly, cheap. Avoids Kimi K2-Thinking (overkill + slow for read-only mapping work).

Fallback chain (in order)

smooth-reasoning (DeepSeek)→
qwen3-coder-plus (Alibaba, 1M ctx)→
claude-haiku-4-5 (Anthropic)

Every primary also has a per-lab variant (e.g. smooth-coding-qwen, smooth-fast-gemini) so callers can pin a specific lab without the cross-vendor chain — useful for compliance, A/B testing, or sticky cache locality.

Why route through Smoo AI

Unified billing

One invoice across every lab. Tier-based token allowances, per-org metering, Stripe-synced overage.

Org-scoped virtual keys

Each organization gets its own key with optional model allowlist and budget cap. Rotate from the dashboard — no downtime.

Cross-lab fallback chains

Every model has a typed fallback chain spanning at least three labs. A Vertex 503, an Anthropic 429, or a DeepSeek timeout degrades to the next provider before surfacing an error.

Drop-in compatibility

OpenAI SDK, LangChain, LlamaIndex, Vercel AI SDK — anything that takes a base URL works unchanged.

Streaming, tool use, JSON mode

Everything the upstream model supports passes through untouched, plus kwargs the OpenAI shape does not cover.

Live OpenAPI spec

Full interactive reference below. The spec tracks upstream provider capabilities in real time.

Interactive API reference

Full endpoint + schema reference for everything the gateway serves — chat completions, embeddings, moderations, and key management. Try every endpoint inline.

GET/models

Model List

Use `/model/info` - to get detailed model information, example - pricing, mode, etc. This is just for compatibility with openai projects like aider. Query Parameters: - include_metadata: Include additional metadata in the response with fallback information - fallback_type: Type of fallbacks to include ("general", "context_window", "content_policy") Defaults to "general" when include_metadata=true - scope: Optional scope parameter. Currently only accepts "expand". When scope=expand is passed, proxy admins, team admins, and org admins will receive all proxy models as if they are a proxy admin.

Requires authentication

Query Parameters

Name	Type	Description
`return_wildcard_routes`	`string`
`team_id`	`string`
`include_model_access_groups`	`string`
`only_model_access_groups`	`string`
`include_metadata`	`string`
`fallback_type`	`string`
`scope`	`string`

Responses

200Successful Response

422Validation Error

Property	Type	Description
`detail`	`array`

Code Examples

curl -X GET https://llm.smoo.ai/models \
  -H "Authorization: Bearer YOUR_ACCESS_TOKEN"