Skip to main content
Mastra 1.0 is available 🎉 Read announcement

Kilo Gateway logoKilo Gateway

Access 263 Kilo Gateway models through Mastra's model router. Authentication is handled automatically using the KILO_API_KEY environment variable.

Learn more in the Kilo Gateway documentation.

.env
KILO_API_KEY=your-api-key
src/mastra/agents/my-agent.ts
import { Agent } from "@mastra/core/agent";

const agent = new Agent({
id: "my-agent",
name: "My Agent",
instructions: "You are a helpful assistant",
model: "kilo/allenai/molmo-2-8b"
});

// Generate a response
const response = await agent.generate("Hello!");

// Stream a response
const stream = await agent.stream("Tell me a story");
for await (const chunk of stream) {
console.log(chunk);
}
info

Mastra uses the OpenAI-compatible /chat/completions endpoint. Some provider-specific features may not be available. Check the Kilo Gateway documentation for details.

Models
Direct link to Models

ModelContextToolsReasoningImageAudioVideoInput $/1MOutput $/1M
kilo/allenai/molmo-2-8b37K$0.20$0.20
kilo/amazon/nova-2-lite-v11.0M$0.30$3
kilo/amazon/nova-pro-v1300K$0.80$3
kilo/anthropic/claude-3-haiku200K$0.25$1
kilo/anthropic/claude-3.5-haiku200K$0.80$4
kilo/anthropic/claude-3.5-sonnet200K$6$30
kilo/anthropic/claude-3.7-sonnet200K$3$15
kilo/anthropic/claude-3.7-sonnet:thinking200K$3$15
kilo/anthropic/claude-haiku-4.5200K$1$5
kilo/anthropic/claude-opus-4200K$15$75
kilo/anthropic/claude-opus-4.1200K$15$75
kilo/anthropic/claude-opus-4.5200K$5$25
kilo/anthropic/claude-opus-4.61.0M$5$25
kilo/anthropic/claude-sonnet-41.0M$3$15
kilo/anthropic/claude-sonnet-4.51.0M$3$15
kilo/arcee-ai/trinity-large-preview:free131K——
kilo/arcee-ai/trinity-mini131K$0.04$0.15
kilo/arcee-ai/trinity-mini:free131K——
kilo/baidu/ernie-4.5-21b-a3b120K$0.07$0.28
kilo/baidu/ernie-4.5-21b-a3b-thinking131K$0.07$0.28
kilo/baidu/ernie-4.5-300b-a47b123K$0.28$1
kilo/baidu/ernie-4.5-vl-28b-a3b30K$0.14$0.56
kilo/baidu/ernie-4.5-vl-424b-a47b123K$0.42$1
kilo/bytedance-seed/seed-1.6262K$0.25$2
kilo/cognitivecomputations/dolphin-mistral-24b-venice-edition:free33K——
kilo/cohere/command-a256K$3$10
kilo/cohere/command-r-08-2024128K$0.15$0.60
kilo/cohere/command-r-plus-08-2024128K$3$10
kilo/cohere/command-r7b-12-2024128K$0.04$0.15
kilo/deepseek/deepseek-chat164K$0.30$1
kilo/deepseek/deepseek-chat-v3-0324164K$0.19$0.87
kilo/deepseek/deepseek-chat-v3.133K$0.15$0.75
kilo/deepseek/deepseek-r164K$0.70$3
kilo/deepseek/deepseek-r1-0528164K$0.40$2
kilo/deepseek/deepseek-r1-0528:free164K——
kilo/deepseek/deepseek-r1-distill-llama-70b131K$0.03$0.11
kilo/deepseek/deepseek-r1-distill-qwen-32b33K$0.29$0.29
kilo/deepseek/deepseek-v3.1-terminus164K$0.21$0.79
kilo/deepseek/deepseek-v3.1-terminus:exacto164K$0.21$0.79
kilo/deepseek/deepseek-v3.2164K$0.25$0.38
kilo/deepseek/deepseek-v3.2-exp164K$0.27$0.41
kilo/deepseek/deepseek-v3.2-speciale164K$0.27$0.41
kilo/essentialai/rnj-1-instruct33K$0.15$0.15
kilo/google/gemini-2.0-flash-0011.0M$0.10$0.40
kilo/google/gemini-2.0-flash-lite-0011.0M$0.07$0.30
kilo/google/gemini-2.5-flash1.0M$0.30$3
kilo/google/gemini-2.5-flash-lite1.0M$0.10$0.40
kilo/google/gemini-2.5-flash-lite-preview-09-20251.0M$0.10$0.40
kilo/google/gemini-2.5-flash-preview-09-20251.0M$0.30$3
kilo/google/gemini-2.5-pro1.0M$1$10
kilo/google/gemini-2.5-pro-preview1.0M$1$10
kilo/google/gemini-2.5-pro-preview-05-061.0M$1$10
kilo/google/gemini-3-flash-preview1.0M$0.50$3
kilo/google/gemini-3-pro-preview1.0M$2$12
kilo/google/gemma-2-27b-it8K$0.65$0.65
kilo/google/gemma-2-9b-it8K$0.03$0.09
kilo/google/gemma-3-12b-it131K$0.03$0.10
kilo/google/gemma-3-12b-it:free33K——
kilo/google/gemma-3-27b-it128K$0.04$0.15
kilo/google/gemma-3-27b-it:free131K——
kilo/google/gemma-3-4b-it96K$0.02$0.07
kilo/google/gemma-3-4b-it:free33K——
kilo/google/gemma-3n-e2b-it:free8K——
kilo/google/gemma-3n-e4b-it33K$0.02$0.04
kilo/google/gemma-3n-e4b-it:free8K——
kilo/gryphe/mythomax-l2-13b4K$0.06$0.06
kilo/inception/mercury128K$0.25$1
kilo/inception/mercury-coder128K$0.25$1
kilo/kilo/auto200K$1$1
kilo/kwaipilot/kat-coder-pro256K$0.21$0.83
kilo/liquid/lfm-2.5-1.2b-instruct:free33K——
kilo/liquid/lfm-2.5-1.2b-thinking:free33K——
kilo/meituan/longcat-flash-chat131K$0.20$0.80
kilo/meta-llama/llama-3-70b-instruct8K$0.51$0.74
kilo/meta-llama/llama-3-8b-instruct8K$0.03$0.04
kilo/meta-llama/llama-3.1-405b-instruct131K$4$4
kilo/meta-llama/llama-3.1-70b-instruct131K$0.40$0.40
kilo/meta-llama/llama-3.1-8b-instruct16K$0.02$0.05
kilo/meta-llama/llama-3.2-11b-vision-instruct131K$0.05$0.05
kilo/meta-llama/llama-3.2-1b-instruct60K$0.03$0.20
kilo/meta-llama/llama-3.2-3b-instruct131K$0.02$0.02
kilo/meta-llama/llama-3.2-3b-instruct:free131K——
kilo/meta-llama/llama-3.3-70b-instruct131K$0.10$0.32
kilo/meta-llama/llama-3.3-70b-instruct:free128K——
kilo/meta-llama/llama-4-maverick1.0M$0.15$0.60
kilo/meta-llama/llama-4-scout328K$0.08$0.30
kilo/meta-llama/llama-guard-3-8b131K$0.02$0.06
kilo/meta-llama/llama-guard-4-12b164K$0.18$0.18
kilo/microsoft/phi-416K$0.06$0.14
kilo/microsoft/wizardlm-2-8x22b66K$0.62$0.62
kilo/minimax/minimax-011.0M$0.20$1
kilo/minimax/minimax-m11.0M$0.40$2
kilo/minimax/minimax-m2197K$0.26$1
kilo/minimax/minimax-m2.1197K$0.27$0.95
kilo/minimax/minimax-m2.5197K$0.30$1
kilo/minimax/minimax-m2.5:free205K——
kilo/mistralai/codestral-2508256K$0.30$0.90
kilo/mistralai/devstral-2512262K$0.05$0.22
kilo/mistralai/devstral-medium131K$0.40$2
kilo/mistralai/devstral-small131K$0.10$0.30
kilo/mistralai/ministral-14b-2512262K$0.20$0.20
kilo/mistralai/mistral-7b-instruct33K$0.20$0.20
kilo/mistralai/mistral-7b-instruct-v0.13K$0.11$0.19
kilo/mistralai/mistral-7b-instruct-v0.333K$0.20$0.20
kilo/mistralai/mistral-large128K$2$6
kilo/mistralai/mistral-large-2411131K$2$6
kilo/mistralai/mistral-large-2512262K$0.50$2
kilo/mistralai/mistral-medium-3131K$0.40$2
kilo/mistralai/mistral-medium-3.1131K$0.40$2
kilo/mistralai/mistral-nemo131K$0.02$0.04
kilo/mistralai/mistral-small-24b-instruct-250133K$0.05$0.08
kilo/mistralai/mistral-small-3.1-24b-instruct131K$0.03$0.11
kilo/mistralai/mistral-small-3.1-24b-instruct:free128K——
kilo/mistralai/mistral-small-3.2-24b-instruct131K$0.06$0.18
kilo/mistralai/mixtral-8x22b-instruct66K$2$6
kilo/mistralai/voxtral-small-24b-250732K$0.10$0.30
kilo/moonshotai/kimi-k2131K$0.50$2
kilo/moonshotai/kimi-k2-0905131K$0.40$2
kilo/moonshotai/kimi-k2-0905:exacto262K$0.60$3
kilo/moonshotai/kimi-k2-thinking262K$0.40$2
kilo/moonshotai/kimi-k2.5262K$0.23$3
kilo/morph/morph-v3-fast82K$0.80$1
kilo/morph/morph-v3-large262K$0.90$2
kilo/nex-agi/deepseek-v3.1-nex-n1131K$0.27$1
kilo/nousresearch/deephermes-3-mistral-24b-preview33K$0.02$0.10
kilo/nousresearch/hermes-2-pro-llama-3-8b8K$0.14$0.14
kilo/nousresearch/hermes-3-llama-3.1-405b131K$1$1
kilo/nousresearch/hermes-3-llama-3.1-405b:free131K——
kilo/nousresearch/hermes-4-405b131K$1$3
kilo/nousresearch/hermes-4-70b131K$0.11$0.38
kilo/nvidia/llama-3.1-nemotron-70b-instruct131K$1$1
kilo/nvidia/llama-3.1-nemotron-ultra-253b-v1131K$0.60$2
kilo/nvidia/llama-3.3-nemotron-super-49b-v1.5131K$0.10$0.40
kilo/nvidia/nemotron-3-nano-30b-a3b262K$0.05$0.20
kilo/nvidia/nemotron-3-nano-30b-a3b:free256K——
kilo/nvidia/nemotron-nano-12b-v2-vl131K$0.20$0.60
kilo/nvidia/nemotron-nano-12b-v2-vl:free128K——
kilo/nvidia/nemotron-nano-9b-v2131K$0.04$0.16
kilo/nvidia/nemotron-nano-9b-v2:free128K——
kilo/openai/chatgpt-4o-latest128K$5$15
kilo/openai/gpt-3.5-turbo16K$0.50$2
kilo/openai/gpt-3.5-turbo-06134K$1$2
kilo/openai/gpt-3.5-turbo-instruct4K$2$2
kilo/openai/gpt-48K$30$60
kilo/openai/gpt-4-turbo128K$10$30
kilo/openai/gpt-4.11.0M$2$8
kilo/openai/gpt-4.1-mini1.0M$0.40$2
kilo/openai/gpt-4.1-nano1.0M$0.10$0.40
kilo/openai/gpt-4o128K$3$10
kilo/openai/gpt-4o-2024-05-13128K$5$15
kilo/openai/gpt-4o-2024-08-06128K$3$10
kilo/openai/gpt-4o-2024-11-20128K$3$10
kilo/openai/gpt-4o-mini128K$0.15$0.60
kilo/openai/gpt-4o-mini-search-preview128K$0.15$0.60
kilo/openai/gpt-4o:extended128K$6$18
kilo/openai/gpt-5400K$1$10
kilo/openai/gpt-5-chat128K$1$10
kilo/openai/gpt-5-codex400K$1$10
kilo/openai/gpt-5-mini400K$0.25$2
kilo/openai/gpt-5-nano400K$0.05$0.40
kilo/openai/gpt-5-pro400K$15$120
kilo/openai/gpt-5.1400K$1$10
kilo/openai/gpt-5.1-chat128K$1$10
kilo/openai/gpt-5.1-codex400K$1$10
kilo/openai/gpt-5.1-codex-max400K$1$10
kilo/openai/gpt-5.1-codex-mini400K$0.25$2
kilo/openai/gpt-5.2400K$2$14
kilo/openai/gpt-5.2-chat128K$2$14
kilo/openai/gpt-5.2-codex400K$2$14
kilo/openai/gpt-5.2-pro400K$21$168
kilo/openai/gpt-oss-120b131K$0.04$0.19
kilo/openai/gpt-oss-120b:exacto131K$0.04$0.19
kilo/openai/gpt-oss-120b:free131K——
kilo/openai/gpt-oss-20b131K$0.03$0.14
kilo/openai/gpt-oss-20b:free131K——
kilo/openai/gpt-oss-safeguard-20b131K$0.07$0.30
kilo/openai/o1200K$15$60
kilo/openai/o1-pro200K$150$600
kilo/openai/o3200K$2$8
kilo/openai/o3-deep-research200K$10$40
kilo/openai/o3-mini200K$1$4
kilo/openai/o3-mini-high200K$1$4
kilo/openai/o3-pro200K$20$80
kilo/openai/o4-mini200K$1$4
kilo/openai/o4-mini-deep-research200K$2$8
kilo/openrouter/aurora-alpha128K——
kilo/perplexity/sonar127K$1$1
kilo/perplexity/sonar-deep-research128K$2$8
kilo/perplexity/sonar-pro200K$3$15
kilo/perplexity/sonar-reasoning-pro128K$2$8
kilo/prime-intellect/intellect-3131K$0.20$1
kilo/qwen/qwen-2.5-72b-instruct33K$0.12$0.39
kilo/qwen/qwen-2.5-7b-instruct33K$0.04$0.10
kilo/qwen/qwen-2.5-coder-32b-instruct33K$0.03$0.11
kilo/qwen/qwen-2.5-vl-7b-instruct33K$0.20$0.20
kilo/qwen/qwen-max33K$2$6
kilo/qwen/qwen-plus1.0M$0.40$1
kilo/qwen/qwen-turbo131K$0.05$0.20
kilo/qwen/qwen-vl-max131K$0.80$3
kilo/qwen/qwen-vl-plus131K$0.21$0.63
kilo/qwen/qwen2.5-coder-7b-instruct33K$0.03$0.09
kilo/qwen/qwen2.5-vl-32b-instruct16K$0.05$0.22
kilo/qwen/qwen2.5-vl-72b-instruct33K$0.15$0.60
kilo/qwen/qwen3-14b41K$0.05$0.22
kilo/qwen/qwen3-235b-a22b41K$0.30$1
kilo/qwen/qwen3-235b-a22b-2507262K$0.07$0.10
kilo/qwen/qwen3-235b-a22b-thinking-2507131K——
kilo/qwen/qwen3-30b-a3b41K$0.06$0.22
kilo/qwen/qwen3-30b-a3b-instruct-2507262K$0.08$0.33
kilo/qwen/qwen3-30b-a3b-thinking-250733K$0.05$0.34
kilo/qwen/qwen3-32b41K$0.08$0.24
kilo/qwen/qwen3-4b131K$0.07$0.27
kilo/qwen/qwen3-4b:free41K——
kilo/qwen/qwen3-8b32K$0.05$0.40
kilo/qwen/qwen3-coder262K$0.22$1
kilo/qwen/qwen3-coder-30b-a3b-instruct160K$0.07$0.27
kilo/qwen/qwen3-coder-flash1.0M$0.30$2
kilo/qwen/qwen3-coder-next262K$0.07$0.30
kilo/qwen/qwen3-coder-plus1.0M$1$5
kilo/qwen/qwen3-coder:exacto262K$0.22$2
kilo/qwen/qwen3-coder:free262K——
kilo/qwen/qwen3-max262K$1$6
kilo/qwen/qwen3-max-thinking262K$1$6
kilo/qwen/qwen3-next-80b-a3b-instruct262K$0.09$1
kilo/qwen/qwen3-next-80b-a3b-instruct:free262K——
kilo/qwen/qwen3-next-80b-a3b-thinking128K$0.15$1
kilo/qwen/qwen3-vl-235b-a22b-instruct262K$0.20$0.88
kilo/qwen/qwen3-vl-235b-a22b-thinking131K——
kilo/qwen/qwen3-vl-30b-a3b-instruct131K$0.13$0.52
kilo/qwen/qwen3-vl-30b-a3b-thinking131K——
kilo/qwen/qwen3-vl-32b-instruct131K$0.10$0.42
kilo/qwen/qwen3-vl-8b-instruct131K$0.08$0.50
kilo/qwen/qwen3-vl-8b-thinking131K$0.12$1
kilo/qwen/qwen3.5-397b-a17b256K$0.60$4
kilo/qwen/qwen3.5-plus-02-151.0M$0.40$2
kilo/qwen/qwq-32b33K$0.15$0.40
kilo/stepfun/step-3.5-flash256K$0.10$0.30
kilo/stepfun/step-3.5-flash:free256K——
kilo/tencent/hunyuan-a13b-instruct131K$0.14$0.57
kilo/tngtech/deepseek-r1t-chimera164K$0.30$1
kilo/tngtech/deepseek-r1t2-chimera164K$0.25$0.85
kilo/tngtech/tng-r1t-chimera164K$0.25$0.85
kilo/writer/palmyra-x51.0M$0.60$6
kilo/x-ai/grok-3131K$3$15
kilo/x-ai/grok-3-beta131K$3$15
kilo/x-ai/grok-3-mini131K$0.30$0.50
kilo/x-ai/grok-3-mini-beta131K$0.30$0.50
kilo/x-ai/grok-4256K$3$15
kilo/x-ai/grok-4-fast2.0M$0.20$0.50
kilo/x-ai/grok-4.1-fast2.0M$0.20$0.50
kilo/x-ai/grok-code-fast-1256K$0.20$2
kilo/xiaomi/mimo-v2-flash262K$0.09$0.29
kilo/z-ai/glm-4.5131K$0.35$2
kilo/z-ai/glm-4.5-air131K$0.13$0.85
kilo/z-ai/glm-4.5-air:free131K——
kilo/z-ai/glm-4.5v66K$0.60$2
kilo/z-ai/glm-4.6203K$0.35$2
kilo/z-ai/glm-4.6:exacto205K$0.44$2
kilo/z-ai/glm-4.6v131K$0.30$0.90
kilo/z-ai/glm-4.7203K$0.40$2
kilo/z-ai/glm-4.7-flash203K$0.06$0.40
kilo/z-ai/glm-5205K$0.30$3
kilo/z-ai/glm-5:free203K——
263 available models

Advanced Configuration
Direct link to Advanced Configuration

Custom Headers
Direct link to Custom Headers

src/mastra/agents/my-agent.ts
const agent = new Agent({
id: "custom-agent",
name: "custom-agent",
model: {
url: "https://api.kilo.ai/api/gateway",
id: "kilo/allenai/molmo-2-8b",
apiKey: process.env.KILO_API_KEY,
headers: {
"X-Custom-Header": "value"
}
}
});

Dynamic Model Selection
Direct link to Dynamic Model Selection

src/mastra/agents/my-agent.ts
const agent = new Agent({
id: "dynamic-agent",
name: "Dynamic Agent",
model: ({ requestContext }) => {
const useAdvanced = requestContext.task === "complex";
return useAdvanced
? "kilo/z-ai/glm-5:free"
: "kilo/allenai/molmo-2-8b";
}
});
On this page