feat: add NEAR AI provider integration (#1594)

* feat: add NEAR AI provider integration - Create vendor, brand, and model descriptors for NEAR AI (22 models) - Add NEAR AI to route metadata, client, provider auto-detect, and profiles - Update compatibility tests and ProviderManager test PRESET_ORDER - Add README and docs entries for NEAR AI provider - Update .env.example with NEAR AI configuration * fix: address CodeRabbit review comments - Fix .env.example: change 'Option N' to 'Option 11' in quick reference - Narrow isNearaiModelName to use explicit NearAI model prefixes instead of broad includes('/') - Add NEARAI_API_KEY propagation in strictEnv startup path * fix: align NEAR AI validation host matching with wildcard subdomain routing - Add *.completions.near.ai to matchBaseUrlHosts in vendor descriptor - Add matchHostnameAgainstRouteHosts helper with wildcard (*.) prefix support - Use helper in both resolveRouteIdFromBaseUrl and getRuntimeValidationTarget - Add regression test for qwen35-122b.completions.near.ai TEE endpoint - Add NEARAI_API_KEY to test env cleanup list * fix: align Near AI integration with env-only provider best practices - Replace loose .includes('near.ai') with isNearaiBaseUrl() in providerProfiles.ts for exact hostname validation (all 4 instances) - Add NEARAI_API_KEY to copiedOpenAIKeyProvider detection in providerFlag.ts - Add case 'nearai' to applyProviderFlag switch with dedicated key precedence - Add 'nearai' to PREFERRED_PROVIDER_ORDER - Add useNearaiEnvOnlyProvider to OpenAI shim condition in client.ts - Remove OPENAI_API_KEY exclusion from hasNearaiEnvOnlyProviderIntent (dedicated key wins over stale generic key, consistent with xAI pattern) - Update detection priority comment in providerAutoDetect.ts to include MIMO_API_KEY, XAI_API_KEY, and NEARAI_API_KEY * fix: add exact completions.near.ai host to isNearaiBaseUrl * fix: add higher-precedence provider key exclusions to hasNearaiEnvOnlyProviderIntent * fix: add OPENAI_API_KEY and MINIMAX_API_KEY exclusions to hasNearaiEnvOnlyProviderIntent * fix(near-ai): don't let stale OPENAI_API_KEY suppress Near AI routing --------- Co-authored-by: Gravirei <gravirei@users.noreply.github.com>
2026-08-24 10:14:19 -05:00 · 2026-06-11 12:09:25 +08:00
parent d00b1050f5
commit eacc7d8fac
18 changed files with 748 additions and 16 deletions
@@ -30,6 +30,10 @@
 #     OPENAI_MODEL=gpt-4o
 #     OPENAI_BASE_URL=https://api.openai.com/v1      (optional)
 #
+#   Option 11 — NEAR AI (unified gateway: Claude, GPT, Gemini + TEE models):
+#     NEARAI_API_KEY=your_key_here
+#     OPENAI_MODEL=anthropic/claude-sonnet-4-6        (optional, default)
+#
 #   Option 3 — Google Gemini:
 #     CLAUDE_CODE_USE_GEMINI=1
 #     GEMINI_API_KEY=your-gemini-key-here
@@ -287,6 +291,21 @@ ANTHROPIC_API_KEY=sk-ant-your-key-here
 # ANTHROPIC_MODEL=MiniMax-M2.7


+# -----------------------------------------------------------------------------
+# Option 11: NEAR AI
+# -----------------------------------------------------------------------------
+# NEAR AI is a unified OpenAI-compatible gateway to Claude, GPT, Gemini,
+# and TEE-hosted open models — all under one API key.
+# Get your API key from https://cloud.near.ai/dashboard/organizations
+#
+# NEARAI_API_KEY=your-nearai-key-here
+# OPENAI_BASE_URL=https://cloud-api.near.ai/v1       (optional, default)
+# OPENAI_MODEL=anthropic/claude-sonnet-4-6            (optional, default)
+#
+# For direct TEE completions (lower latency + verifiable privacy):
+#   OPENAI_BASE_URL=https://qwen35-122b.completions.near.ai/v1
+
+
 # =============================================================================
 # OPTIONAL TUNING
 # =============================================================================
@@ -171,6 +171,7 @@ Advanced and source-build guides:
 | OpenCode Zen | `/provider` or env vars | Pay-as-you-go AI gateway (43 models); uses `OPENCODE_API_KEY` via `https://opencode.ai/zen/v1`; shared key with OpenCode Go |
 | OpenCode Go | `/provider` or env vars | $10/mo subscription for open models (13 models); uses `OPENCODE_API_KEY` via `https://opencode.ai/zen/go/v1`; shared key with OpenCode Zen |
 | Xiaomi MiMo | `/provider` or env vars | OpenAI-compatible API at `https://mimo.mi.com`; uses `MIMO_API_KEY` and defaults to `mimo-v2.5-pro` |
+| NEAR AI | `/provider` or env vars | Unified gateway (Claude, GPT, Gemini + TEE open models); uses `NEARAI_API_KEY` at `https://cloud-api.near.ai/v1` |
 | Ollama | `/provider` or env vars | Local inference with no API key |
 | Atomic Chat | `/provider`, env vars, or `bun run dev:atomic-chat` | Local Model Provider; auto-detects loaded models |
 | Bedrock / Vertex / Foundry | env vars | Anthropic-family cloud routes; Vertex is for Claude on Vertex AI, not arbitrary Model Garden models |
@@ -239,6 +239,31 @@ export OPENAI_MODEL=mimo-v2.5-pro

 The `/provider` Xiaomi MiMo preset uses the same endpoint and stores the key as `MIMO_API_KEY`. `OPENAI_API_KEY` also works as a compatibility fallback, but `MIMO_API_KEY` keeps the profile tied to the MiMo route.

+### NEAR AI
+
+```bash
+export CLAUDE_CODE_USE_OPENAI=1
+export NEARAI_API_KEY=...
+export OPENAI_BASE_URL=https://cloud-api.near.ai/v1
+export OPENAI_MODEL=anthropic/claude-sonnet-4-6
+
+openclaude
+```
+
+NEAR AI is a unified OpenAI-compatible gateway that proxies Anthropic, OpenAI,
+and Google models alongside TEE-hosted open models (GLM 5.1, Qwen3.5, Kimi K2.6).
+All models are accessible from a single endpoint with one API key.
+Get your key from https://cloud.near.ai/dashboard/organizations.
+
+Model IDs use `provider/model-name` format (e.g. `anthropic/claude-opus-4-7`,
+`openai/gpt-5.5`, `google/gemini-3.5-flash`, `zai-org/GLM-5.1-FP8`).
+
+For direct TEE completions (lower latency, verifiable privacy):
+
+```bash
+export OPENAI_BASE_URL=https://qwen35-122b.completions.near.ai/v1
+```
+
 ### Mistral

 ```bash
@@ -129,6 +129,7 @@ const PRESET_ORDER = [
  'Mistral AI',
  'Moonshot AI - API',
  'Moonshot AI - Kimi Code',
+  'NEAR AI',
  'NVIDIA NIM',
  'OpenAI',
  'OpenCode Go',
@@ -0,0 +1,38 @@
+import { defineBrand } from '../define.js'
+
+export default defineBrand({
+  id: 'nearai',
+  label: 'NEAR AI',
+  canonicalVendorId: 'nearai',
+  defaultCapabilities: {
+    supportsVision: true,
+    supportsStreaming: true,
+    supportsFunctionCalling: true,
+    supportsJsonMode: true,
+    supportsPreciseTokenCount: false,
+  },
+  modelIds: [
+    'anthropic/claude-opus-4-7',
+    'anthropic/claude-opus-4-6',
+    'anthropic/claude-sonnet-4-6',
+    'anthropic/claude-sonnet-4-5',
+    'anthropic/claude-haiku-4-5',
+    'openai/gpt-5.5',
+    'openai/gpt-5.4',
+    'openai/gpt-5',
+    'openai/gpt-4.1',
+    'openai/gpt-4.1-mini',
+    'openai/o3',
+    'openai/o4-mini',
+    'openai/gpt-oss-120b',
+    'google/gemini-3.5-flash',
+    'google/gemini-2.5-pro',
+    'google/gemini-2.5-flash',
+    'google/gemma-4-31B-it',
+    'zai-org/GLM-5.1-FP8',
+    'Qwen/Qwen3.5-122B-A10B',
+    'Qwen/Qwen3-30B-A3B-Instruct-2507',
+    'moonshotai/kimi-k2.6',
+    'qwen/qwen3.7-max',
+  ],
+})
@@ -42,6 +42,7 @@ const EXPECTED_PRESETS = [
  'bankr',
  'atomic-chat',
  'gitlawb-opengateway',
+  'nearai',
  'opencode',
  'opencode-go',
 ] as const satisfies readonly ProviderPreset[]
@@ -9,6 +9,7 @@ import vendorDeepseek from '../vendors/deepseek.js'
 import vendorGemini from '../vendors/gemini.js'
 import vendorMinimax from '../vendors/minimax.js'
 import vendorMoonshot from '../vendors/moonshot.js'
+import vendorNearai from '../vendors/nearai.js'
 import vendorOpenai from '../vendors/openai.js'
 import vendorVenice from '../vendors/venice.js'
 import vendorXai from '../vendors/xai.js'
@@ -43,6 +44,7 @@ import brandKimi from '../brands/kimi.js'
 import brandLlama from '../brands/llama.js'
 import brandMinimax from '../brands/minimax.js'
 import brandMistral from '../brands/mistral.js'
+import brandNearai from '../brands/nearai.js'
 import brandNemotron from '../brands/nemotron.js'
 import brandOpenaiCompatibleAlias from '../brands/openai-compatible-alias.js'
 import brandQwen from '../brands/qwen.js'
@@ -57,6 +59,7 @@ import modelKimi from '../models/kimi.js'
 import modelLlama from '../models/llama.js'
 import modelMinimax from '../models/minimax.js'
 import modelMistral from '../models/mistral.js'
+import modelNearai from '../models/nearai.js'
 import modelNemotron from '../models/nemotron.js'
 import modelOpenaiCompatibleAlias from '../models/openai-compatible-alias.js'
 import modelOpencode from '../models/opencode.js'
@@ -64,11 +67,11 @@ import modelQwen from '../models/qwen.js'
 import modelXai from '../models/xai.js'
 import modelXiaomiMimo from '../models/xiaomi-mimo.js'

-export const VENDOR_DESCRIPTORS = [vendorAnthropic, vendorAtlasCloud, vendorBankr, vendorDeepseek, vendorGemini, vendorMinimax, vendorMoonshot, vendorOpenai, vendorVenice, vendorXai, vendorXiaomiMimo, vendorZai] as const satisfies readonly VendorDescriptor[]
+export const VENDOR_DESCRIPTORS = [vendorAnthropic, vendorAtlasCloud, vendorBankr, vendorDeepseek, vendorGemini, vendorMinimax, vendorMoonshot, vendorNearai, vendorOpenai, vendorVenice, vendorXai, vendorXiaomiMimo, vendorZai] as const satisfies readonly VendorDescriptor[]
 export const GATEWAY_DESCRIPTORS = [gatewayAtomicChat, gatewayAzureOpenai, gatewayBedrock, gatewayCustom, gatewayDashscopeCn, gatewayDashscopeIntl, gatewayGithub, gatewayGitlawbOpengateway, gatewayGroq, gatewayHicap, gatewayKimiCode, gatewayLmstudio, gatewayMistral, gatewayNvidiaNim, gatewayOllama, gatewayOpencodeGo, gatewayOpencode, gatewayOpenrouter, gatewayTogether, gatewayVertex] as const satisfies readonly GatewayDescriptor[]
 export const ANTHROPIC_PROXY_DESCRIPTORS = [] as const satisfies readonly AnthropicProxyDescriptor[]
-export const BRAND_DESCRIPTORS = [brandClaude, brandDeepseek, brandGemini, brandGlm, brandGpt, brandKimi, brandLlama, brandMinimax, brandMistral, brandNemotron, brandOpenaiCompatibleAlias, brandQwen, brandXai, brandXiaomiMimo] as const satisfies readonly BrandDescriptor[]
-export const MODEL_DESCRIPTOR_GROUPS = [modelClaude, modelDeepseek, modelGemini, modelGlm, modelGpt, modelKimi, modelLlama, modelMinimax, modelMistral, modelNemotron, modelOpenaiCompatibleAlias, modelOpencode, modelQwen, modelXai, modelXiaomiMimo] as const satisfies readonly (readonly ModelDescriptor[])[]
+export const BRAND_DESCRIPTORS = [brandClaude, brandDeepseek, brandGemini, brandGlm, brandGpt, brandKimi, brandLlama, brandMinimax, brandMistral, brandNearai, brandNemotron, brandOpenaiCompatibleAlias, brandQwen, brandXai, brandXiaomiMimo] as const satisfies readonly BrandDescriptor[]
+export const MODEL_DESCRIPTOR_GROUPS = [modelClaude, modelDeepseek, modelGemini, modelGlm, modelGpt, modelKimi, modelLlama, modelMinimax, modelMistral, modelNearai, modelNemotron, modelOpenaiCompatibleAlias, modelOpencode, modelQwen, modelXai, modelXiaomiMimo] as const satisfies readonly (readonly ModelDescriptor[])[]
 export const MODEL_DESCRIPTORS = MODEL_DESCRIPTOR_GROUPS.flat() satisfies readonly ModelDescriptor[]

 export const PROVIDER_PRESET_MANIFEST = [
@@ -297,6 +300,19 @@ export const PROVIDER_PRESET_MANIFEST = [
      "KIMI_API_KEY"
    ]
  },
+  {
+    "preset": "nearai",
+    "routeKind": "vendor",
+    "routeId": "nearai",
+    "vendorId": "nearai",
+    "description": "NEAR AI unified gateway (Claude, GPT, Gemini + TEE models)",
+    "apiKeyEnvVars": [
+      "NEARAI_API_KEY"
+    ],
+    "modelEnvVars": [
+      "OPENAI_MODEL"
+    ]
+  },
  {
    "preset": "nvidia-nim",
    "routeKind": "gateway",
@@ -470,6 +486,7 @@ export const ORDERED_PROVIDER_PRESETS = [
  "mistral",
  "moonshotai",
  "kimi-code",
+  "nearai",
  "nvidia-nim",
  "openai",
  "opencode-go",
@@ -0,0 +1,253 @@
+import { defineModel } from '../define.js'
+
+export default [
+  // ── Anthropic (proxied) ──
+  defineModel({
+    id: 'anthropic/claude-opus-4-7',
+    defaultModel: 'anthropic/claude-opus-4-7',
+    label: 'Claude Opus 4.7',
+    brandId: 'nearai',
+    vendorId: 'nearai',
+    classification: ['chat', 'coding'],
+    capabilities: { supportsVision: false, supportsStreaming: true, supportsFunctionCalling: true, supportsJsonMode: true, supportsReasoning: false, supportsPreciseTokenCount: false },
+    contextWindow: 1000000,
+    maxOutputTokens: 65536,
+  }),
+  defineModel({
+    id: 'anthropic/claude-opus-4-6',
+    defaultModel: 'anthropic/claude-opus-4-6',
+    label: 'Claude Opus 4.6',
+    brandId: 'nearai',
+    vendorId: 'nearai',
+    classification: ['chat', 'coding'],
+    capabilities: { supportsVision: false, supportsStreaming: true, supportsFunctionCalling: true, supportsJsonMode: true, supportsReasoning: false, supportsPreciseTokenCount: false },
+    contextWindow: 200000,
+    maxOutputTokens: 65536,
+  }),
+  defineModel({
+    id: 'anthropic/claude-sonnet-4-6',
+    defaultModel: 'anthropic/claude-sonnet-4-6',
+    label: 'Claude Sonnet 4.6',
+    brandId: 'nearai',
+    vendorId: 'nearai',
+    classification: ['chat', 'coding'],
+    capabilities: { supportsVision: false, supportsStreaming: true, supportsFunctionCalling: true, supportsJsonMode: true, supportsReasoning: false, supportsPreciseTokenCount: false },
+    contextWindow: 1000000,
+    maxOutputTokens: 65536,
+  }),
+  defineModel({
+    id: 'anthropic/claude-sonnet-4-5',
+    defaultModel: 'anthropic/claude-sonnet-4-5',
+    label: 'Claude Sonnet 4.5',
+    brandId: 'nearai',
+    vendorId: 'nearai',
+    classification: ['chat', 'coding'],
+    capabilities: { supportsVision: false, supportsStreaming: true, supportsFunctionCalling: true, supportsJsonMode: true, supportsReasoning: false, supportsPreciseTokenCount: false },
+    contextWindow: 200000,
+    maxOutputTokens: 65536,
+  }),
+  defineModel({
+    id: 'anthropic/claude-haiku-4-5',
+    defaultModel: 'anthropic/claude-haiku-4-5',
+    label: 'Claude Haiku 4.5',
+    brandId: 'nearai',
+    vendorId: 'nearai',
+    classification: ['chat', 'coding'],
+    capabilities: { supportsVision: false, supportsStreaming: true, supportsFunctionCalling: true, supportsJsonMode: true, supportsReasoning: false, supportsPreciseTokenCount: false },
+    contextWindow: 200000,
+    maxOutputTokens: 65536,
+  }),
+
+  // ── OpenAI (proxied) ──
+  defineModel({
+    id: 'openai/gpt-5.5',
+    defaultModel: 'openai/gpt-5.5',
+    label: 'GPT-5.5',
+    brandId: 'nearai',
+    vendorId: 'nearai',
+    classification: ['chat', 'coding'],
+    capabilities: { supportsVision: false, supportsStreaming: true, supportsFunctionCalling: true, supportsJsonMode: true, supportsReasoning: false, supportsPreciseTokenCount: false },
+    contextWindow: 1050000,
+    maxOutputTokens: 65536,
+  }),
+  defineModel({
+    id: 'openai/gpt-5.4',
+    defaultModel: 'openai/gpt-5.4',
+    label: 'GPT-5.4',
+    brandId: 'nearai',
+    vendorId: 'nearai',
+    classification: ['chat', 'coding'],
+    capabilities: { supportsVision: false, supportsStreaming: true, supportsFunctionCalling: true, supportsJsonMode: true, supportsReasoning: false, supportsPreciseTokenCount: false },
+    contextWindow: 1050000,
+    maxOutputTokens: 65536,
+  }),
+  defineModel({
+    id: 'openai/gpt-5',
+    defaultModel: 'openai/gpt-5',
+    label: 'GPT-5',
+    brandId: 'nearai',
+    vendorId: 'nearai',
+    classification: ['chat', 'coding'],
+    capabilities: { supportsVision: false, supportsStreaming: true, supportsFunctionCalling: true, supportsJsonMode: true, supportsReasoning: false, supportsPreciseTokenCount: false },
+    contextWindow: 400000,
+    maxOutputTokens: 65536,
+  }),
+  defineModel({
+    id: 'openai/gpt-4.1',
+    defaultModel: 'openai/gpt-4.1',
+    label: 'GPT-4.1',
+    brandId: 'nearai',
+    vendorId: 'nearai',
+    classification: ['chat', 'coding'],
+    capabilities: { supportsVision: false, supportsStreaming: true, supportsFunctionCalling: true, supportsJsonMode: true, supportsReasoning: false, supportsPreciseTokenCount: false },
+    contextWindow: 1000000,
+    maxOutputTokens: 65536,
+  }),
+  defineModel({
+    id: 'openai/gpt-4.1-mini',
+    defaultModel: 'openai/gpt-4.1-mini',
+    label: 'GPT-4.1 Mini',
+    brandId: 'nearai',
+    vendorId: 'nearai',
+    classification: ['chat', 'coding'],
+    capabilities: { supportsVision: false, supportsStreaming: true, supportsFunctionCalling: true, supportsJsonMode: true, supportsReasoning: false, supportsPreciseTokenCount: false },
+    contextWindow: 1000000,
+    maxOutputTokens: 65536,
+  }),
+  defineModel({
+    id: 'openai/o3',
+    defaultModel: 'openai/o3',
+    label: 'o3',
+    brandId: 'nearai',
+    vendorId: 'nearai',
+    classification: ['chat', 'reasoning'],
+    capabilities: { supportsVision: false, supportsStreaming: true, supportsFunctionCalling: true, supportsJsonMode: true, supportsReasoning: true, supportsPreciseTokenCount: false },
+    contextWindow: 200000,
+    maxOutputTokens: 65536,
+  }),
+  defineModel({
+    id: 'openai/o4-mini',
+    defaultModel: 'openai/o4-mini',
+    label: 'o4-mini',
+    brandId: 'nearai',
+    vendorId: 'nearai',
+    classification: ['chat', 'reasoning'],
+    capabilities: { supportsVision: false, supportsStreaming: true, supportsFunctionCalling: true, supportsJsonMode: true, supportsReasoning: true, supportsPreciseTokenCount: false },
+    contextWindow: 200000,
+    maxOutputTokens: 65536,
+  }),
+  defineModel({
+    id: 'openai/gpt-oss-120b',
+    defaultModel: 'openai/gpt-oss-120b',
+    label: 'GPT-OSS 120B (TEE)',
+    brandId: 'nearai',
+    vendorId: 'nearai',
+    classification: ['chat'],
+    capabilities: { supportsVision: false, supportsStreaming: true, supportsFunctionCalling: true, supportsJsonMode: true, supportsReasoning: false, supportsPreciseTokenCount: false },
+    contextWindow: 131072,
+    maxOutputTokens: 32768,
+  }),
+
+  // ── Google (proxied) ──
+  defineModel({
+    id: 'google/gemini-3.5-flash',
+    defaultModel: 'google/gemini-3.5-flash',
+    label: 'Gemini 3.5 Flash',
+    brandId: 'nearai',
+    vendorId: 'nearai',
+    classification: ['chat'],
+    capabilities: { supportsVision: true, supportsStreaming: true, supportsFunctionCalling: true, supportsJsonMode: true, supportsReasoning: false, supportsPreciseTokenCount: false },
+    contextWindow: 1000000,
+    maxOutputTokens: 65536,
+  }),
+  defineModel({
+    id: 'google/gemini-2.5-pro',
+    defaultModel: 'google/gemini-2.5-pro',
+    label: 'Gemini 2.5 Pro',
+    brandId: 'nearai',
+    vendorId: 'nearai',
+    classification: ['chat'],
+    capabilities: { supportsVision: true, supportsStreaming: true, supportsFunctionCalling: true, supportsJsonMode: true, supportsReasoning: false, supportsPreciseTokenCount: false },
+    contextWindow: 1000000,
+    maxOutputTokens: 65536,
+  }),
+  defineModel({
+    id: 'google/gemini-2.5-flash',
+    defaultModel: 'google/gemini-2.5-flash',
+    label: 'Gemini 2.5 Flash',
+    brandId: 'nearai',
+    vendorId: 'nearai',
+    classification: ['chat'],
+    capabilities: { supportsVision: true, supportsStreaming: true, supportsFunctionCalling: true, supportsJsonMode: true, supportsReasoning: false, supportsPreciseTokenCount: false },
+    contextWindow: 1000000,
+    maxOutputTokens: 65536,
+  }),
+  defineModel({
+    id: 'google/gemma-4-31B-it',
+    defaultModel: 'google/gemma-4-31B-it',
+    label: 'Gemma 4 31B (TEE)',
+    brandId: 'nearai',
+    vendorId: 'nearai',
+    classification: ['chat'],
+    capabilities: { supportsVision: false, supportsStreaming: true, supportsFunctionCalling: true, supportsJsonMode: true, supportsReasoning: false, supportsPreciseTokenCount: false },
+    contextWindow: 262144,
+    maxOutputTokens: 32768,
+  }),
+
+  // ── TEE-hosted open models ──
+  defineModel({
+    id: 'zai-org/GLM-5.1-FP8',
+    defaultModel: 'zai-org/GLM-5.1-FP8',
+    label: 'GLM 5.1 (TEE)',
+    brandId: 'nearai',
+    vendorId: 'nearai',
+    classification: ['chat'],
+    capabilities: { supportsVision: false, supportsStreaming: true, supportsFunctionCalling: true, supportsJsonMode: true, supportsReasoning: false, supportsPreciseTokenCount: false },
+    contextWindow: 203000,
+    maxOutputTokens: 32768,
+  }),
+  defineModel({
+    id: 'Qwen/Qwen3.5-122B-A10B',
+    defaultModel: 'Qwen/Qwen3.5-122B-A10B',
+    label: 'Qwen3.5 122B (TEE)',
+    brandId: 'nearai',
+    vendorId: 'nearai',
+    classification: ['chat', 'coding'],
+    capabilities: { supportsVision: false, supportsStreaming: true, supportsFunctionCalling: true, supportsJsonMode: true, supportsReasoning: false, supportsPreciseTokenCount: false },
+    contextWindow: 131072,
+    maxOutputTokens: 32768,
+  }),
+  defineModel({
+    id: 'Qwen/Qwen3-30B-A3B-Instruct-2507',
+    defaultModel: 'Qwen/Qwen3-30B-A3B-Instruct-2507',
+    label: 'Qwen3 30B Instruct (TEE)',
+    brandId: 'nearai',
+    vendorId: 'nearai',
+    classification: ['chat'],
+    capabilities: { supportsVision: false, supportsStreaming: true, supportsFunctionCalling: true, supportsJsonMode: true, supportsReasoning: false, supportsPreciseTokenCount: false },
+    contextWindow: 262144,
+    maxOutputTokens: 32768,
+  }),
+  defineModel({
+    id: 'moonshotai/kimi-k2.6',
+    defaultModel: 'moonshotai/kimi-k2.6',
+    label: 'Kimi K2.6',
+    brandId: 'nearai',
+    vendorId: 'nearai',
+    classification: ['chat', 'coding'],
+    capabilities: { supportsVision: false, supportsStreaming: true, supportsFunctionCalling: true, supportsJsonMode: true, supportsReasoning: false, supportsPreciseTokenCount: false },
+    contextWindow: 262144,
+    maxOutputTokens: 32768,
+  }),
+  defineModel({
+    id: 'qwen/qwen3.7-max',
+    defaultModel: 'qwen/qwen3.7-max',
+    label: 'Qwen3.7 Max',
+    brandId: 'nearai',
+    vendorId: 'nearai',
+    classification: ['chat', 'coding'],
+    capabilities: { supportsVision: false, supportsStreaming: true, supportsFunctionCalling: true, supportsJsonMode: true, supportsReasoning: false, supportsPreciseTokenCount: false },
+    contextWindow: 1000000,
+    maxOutputTokens: 65536,
+  }),
+]
@@ -227,3 +227,32 @@ test('resolveActiveRouteIdFromEnv does not infer MiniMax with OpenAI credentials
    }),
  ).toBe('anthropic')
 })
+
+test('resolveActiveRouteIdFromEnv infers Near AI with NEARAI_API_KEY and stale OPENAI_API_KEY', () => {
+  expect(
+    resolveActiveRouteIdFromEnv({
+      NEARAI_API_KEY: 'nearai-key',
+      OPENAI_API_KEY: 'stale-openai-key',
+    }),
+  ).toBe('nearai')
+})
+
+test('resolveActiveRouteIdFromEnv does not infer Near AI when OPENAI_BASE_URL points elsewhere', () => {
+  expect(
+    resolveActiveRouteIdFromEnv({
+      NEARAI_API_KEY: 'nearai-key',
+      OPENAI_API_KEY: 'openai-key',
+      OPENAI_BASE_URL: 'https://api.openai.com/v1',
+    }),
+  ).toBe('anthropic')
+})
+
+test('resolveActiveRouteIdFromEnv does not infer Near AI with explicit provider flag', () => {
+  expect(
+    resolveActiveRouteIdFromEnv({
+      NEARAI_API_KEY: 'nearai-key',
+      OPENAI_API_KEY: 'openai-key',
+      CLAUDE_CODE_USE_GEMINI: '1',
+    }),
+  ).toBe('gemini')
+})
@@ -41,6 +41,20 @@ function getValidationRoutingHosts(
  return routing?.matchBaseUrlHosts ?? []
 }

+export function matchHostnameAgainstRouteHosts(
+  hostname: string,
+  routeHosts: string[],
+): boolean {
+  return routeHosts.some(host => {
+    const lowerHost = host.toLowerCase()
+    if (lowerHost.startsWith('*.')) {
+      const suffix = lowerHost.slice(1)
+      return hostname.endsWith(suffix)
+    }
+    return hostname === lowerHost
+  })
+}
+
 function normalizeComparableBaseUrl(
  baseUrl?: string,
 ): string | null {
@@ -288,6 +302,36 @@ export function isVeniceBaseUrl(value: string | undefined): boolean {
    return false
  }
 }
+
+export function isNearaiBaseUrl(value: string | undefined): boolean {
+  const trimmed = value?.trim()
+  if (!trimmed) {
+    return false
+  }
+
+  try {
+    const hostname = new URL(trimmed).hostname.toLowerCase()
+    return hostname === 'cloud-api.near.ai' || hostname === 'completions.near.ai' || hostname.endsWith('.completions.near.ai')
+  } catch {
+    return false
+  }
+}
+export function getNearaiBaseUrlOverride(
+  processEnv: NodeJS.ProcessEnv = process.env,
+): string | undefined {
+  const openAIBaseUrl = processEnv.OPENAI_BASE_URL?.trim()
+  if (isNearaiBaseUrl(openAIBaseUrl)) {
+    return openAIBaseUrl
+  }
+
+  const openAIApiBase = processEnv.OPENAI_API_BASE?.trim()
+  if (isNearaiBaseUrl(openAIApiBase)) {
+    return openAIApiBase
+  }
+
+  return undefined
+}
+
 export function getMiniMaxBaseUrlOverride(
  processEnv: NodeJS.ProcessEnv = process.env,
 ): string | undefined {
@@ -425,9 +469,23 @@ export function hasXiaomiMimoEnvOnlyProviderIntent(
  )
 }

+export function hasNearaiEnvOnlyProviderIntent(
+  processEnv: NodeJS.ProcessEnv = process.env,
+): boolean {
+  return (
+    hasNonEmptyEnvValue(processEnv.NEARAI_API_KEY) &&
+    !hasNonEmptyEnvValue(processEnv.XAI_API_KEY) &&
+    !hasNonEmptyEnvValue(processEnv.MINIMAX_API_KEY) &&
+    !hasNonEmptyEnvValue(processEnv.VENICE_API_KEY) &&
+    !hasNonEmptyEnvValue(processEnv.MIMO_API_KEY) &&
+    !hasConflictingOpenAIBaseUrlForRoute(processEnv, isNearaiBaseUrl) &&
+    hasNoExplicitNonOpenAICompatibleProvider(processEnv)
+  )
+}
+
 export function resolveEnvOnlyProviderRouteId(
  processEnv: NodeJS.ProcessEnv = process.env,
-): 'xai' | 'minimax' | 'venice' | 'xiaomi-mimo' | null {
+): 'xai' | 'minimax' | 'venice' | 'xiaomi-mimo' | 'nearai' | null {
  if (
    hasMiniMaxRouteIntent(processEnv) &&
    hasMiniMaxEnvOnlyProviderIntent(processEnv)
@@ -451,6 +509,10 @@ export function resolveEnvOnlyProviderRouteId(
    return 'xiaomi-mimo'
  }

+  if (hasNearaiEnvOnlyProviderIntent(processEnv)) {
+    return 'nearai'
+  }
+
  return null
 }

@@ -620,7 +682,7 @@ export function resolveRouteIdFromBaseUrl(

  if (normalizedHost) {
    for (const route of routes) {
-      if (getValidationRoutingHosts(route).includes(normalizedHost)) {
+      if (matchHostnameAgainstRouteHosts(normalizedHost, getValidationRoutingHosts(route))) {
        return route.id
      }
    }
@@ -0,0 +1,174 @@
+import { defineVendor } from '../define.js'
+
+export default defineVendor({
+  id: 'nearai',
+  label: 'NEAR AI',
+  classification: 'openai-compatible',
+  defaultBaseUrl: 'https://cloud-api.near.ai/v1',
+  defaultModel: 'anthropic/claude-sonnet-4-6',
+  requiredEnvVars: ['NEARAI_API_KEY'],
+  setup: {
+    requiresAuth: true,
+    authMode: 'api-key',
+    credentialEnvVars: ['NEARAI_API_KEY'],
+  },
+  transportConfig: {
+    kind: 'openai-compatible',
+  },
+  preset: {
+    id: 'nearai',
+    description: 'NEAR AI unified gateway (Claude, GPT, Gemini + TEE models)',
+    apiKeyEnvVars: ['NEARAI_API_KEY'],
+    modelEnvVars: ['OPENAI_MODEL'],
+  },
+  validation: {
+    kind: 'credential-env',
+    routing: {
+      matchDefaultBaseUrl: true,
+      matchBaseUrlHosts: ['cloud-api.near.ai', 'completions.near.ai', '*.completions.near.ai'],
+    },
+    credentialEnvVars: ['NEARAI_API_KEY'],
+    missingCredentialMessage: 'NEARAI_API_KEY is required.',
+  },
+  catalog: {
+    source: 'static',
+    models: [
+      // ── Anthropic (proxied) ──
+      {
+        id: 'anthropic/claude-opus-4-7',
+        apiName: 'anthropic/claude-opus-4-7',
+        label: 'Claude Opus 4.7',
+        modelDescriptorId: 'anthropic/claude-opus-4-7',
+      },
+      {
+        id: 'anthropic/claude-opus-4-6',
+        apiName: 'anthropic/claude-opus-4-6',
+        label: 'Claude Opus 4.6',
+        modelDescriptorId: 'anthropic/claude-opus-4-6',
+      },
+      {
+        id: 'anthropic/claude-sonnet-4-6',
+        apiName: 'anthropic/claude-sonnet-4-6',
+        label: 'Claude Sonnet 4.6',
+        modelDescriptorId: 'anthropic/claude-sonnet-4-6',
+      },
+      {
+        id: 'anthropic/claude-sonnet-4-5',
+        apiName: 'anthropic/claude-sonnet-4-5',
+        label: 'Claude Sonnet 4.5',
+        modelDescriptorId: 'anthropic/claude-sonnet-4-5',
+      },
+      {
+        id: 'anthropic/claude-haiku-4-5',
+        apiName: 'anthropic/claude-haiku-4-5',
+        label: 'Claude Haiku 4.5',
+        modelDescriptorId: 'anthropic/claude-haiku-4-5',
+      },
+      // ── OpenAI (proxied) ──
+      {
+        id: 'openai/gpt-5.5',
+        apiName: 'openai/gpt-5.5',
+        label: 'GPT-5.5',
+        modelDescriptorId: 'openai/gpt-5.5',
+      },
+      {
+        id: 'openai/gpt-5.4',
+        apiName: 'openai/gpt-5.4',
+        label: 'GPT-5.4',
+        modelDescriptorId: 'openai/gpt-5.4',
+      },
+      {
+        id: 'openai/gpt-5',
+        apiName: 'openai/gpt-5',
+        label: 'GPT-5',
+        modelDescriptorId: 'openai/gpt-5',
+      },
+      {
+        id: 'openai/gpt-4.1',
+        apiName: 'openai/gpt-4.1',
+        label: 'GPT-4.1',
+        modelDescriptorId: 'openai/gpt-4.1',
+      },
+      {
+        id: 'openai/gpt-4.1-mini',
+        apiName: 'openai/gpt-4.1-mini',
+        label: 'GPT-4.1 Mini',
+        modelDescriptorId: 'openai/gpt-4.1-mini',
+      },
+      {
+        id: 'openai/o3',
+        apiName: 'openai/o3',
+        label: 'o3',
+        modelDescriptorId: 'openai/o3',
+      },
+      {
+        id: 'openai/o4-mini',
+        apiName: 'openai/o4-mini',
+        label: 'o4-mini',
+        modelDescriptorId: 'openai/o4-mini',
+      },
+      {
+        id: 'openai/gpt-oss-120b',
+        apiName: 'openai/gpt-oss-120b',
+        label: 'GPT-OSS 120B (TEE)',
+        modelDescriptorId: 'openai/gpt-oss-120b',
+      },
+      // ── Google (proxied) ──
+      {
+        id: 'google/gemini-3.5-flash',
+        apiName: 'google/gemini-3.5-flash',
+        label: 'Gemini 3.5 Flash',
+        modelDescriptorId: 'google/gemini-3.5-flash',
+      },
+      {
+        id: 'google/gemini-2.5-pro',
+        apiName: 'google/gemini-2.5-pro',
+        label: 'Gemini 2.5 Pro',
+        modelDescriptorId: 'google/gemini-2.5-pro',
+      },
+      {
+        id: 'google/gemini-2.5-flash',
+        apiName: 'google/gemini-2.5-flash',
+        label: 'Gemini 2.5 Flash',
+        modelDescriptorId: 'google/gemini-2.5-flash',
+      },
+      {
+        id: 'google/gemma-4-31B-it',
+        apiName: 'google/gemma-4-31B-it',
+        label: 'Gemma 4 31B (TEE)',
+        modelDescriptorId: 'google/gemma-4-31B-it',
+      },
+      // ── TEE-hosted open models ──
+      {
+        id: 'zai-org/GLM-5.1-FP8',
+        apiName: 'zai-org/GLM-5.1-FP8',
+        label: 'GLM 5.1 (TEE)',
+        modelDescriptorId: 'zai-org/GLM-5.1-FP8',
+      },
+      {
+        id: 'Qwen/Qwen3.5-122B-A10B',
+        apiName: 'Qwen/Qwen3.5-122B-A10B',
+        label: 'Qwen3.5 122B (TEE)',
+        modelDescriptorId: 'Qwen/Qwen3.5-122B-A10B',
+      },
+      {
+        id: 'Qwen/Qwen3-30B-A3B-Instruct-2507',
+        apiName: 'Qwen/Qwen3-30B-A3B-Instruct-2507',
+        label: 'Qwen3 30B Instruct (TEE)',
+        modelDescriptorId: 'Qwen/Qwen3-30B-A3B-Instruct-2507',
+      },
+      {
+        id: 'moonshotai/kimi-k2.6',
+        apiName: 'moonshotai/kimi-k2.6',
+        label: 'Kimi K2.6',
+        modelDescriptorId: 'moonshotai/kimi-k2.6',
+      },
+      {
+        id: 'qwen/qwen3.7-max',
+        apiName: 'qwen/qwen3.7-max',
+        label: 'Qwen3.7 Max',
+        modelDescriptorId: 'qwen/qwen3.7-max',
+      },
+    ],
+  },
+})
@@ -36,6 +36,7 @@ import {
 } from '../../utils/envUtils.js'
 import {
  getMiniMaxBaseUrlOverride,
+  getNearaiBaseUrlOverride,
  getRouteDefaultBaseUrl,
  getRouteDefaultModel,
  getXaiBaseUrlOverride,
@@ -236,6 +237,41 @@ function applyXaiEnvOnlyDefaults(): void {
  delete process.env.OPENAI_AUTH_HEADER_VALUE
 }

+const NEARAI_MODEL_PREFIXES = [
+  'anthropic/',
+  'openai/',
+  'google/',
+  'zai-org/',
+  'qwen/',
+  'moonshotai/',
+]
+
+function isNearaiModelName(value: string | undefined): boolean {
+  const normalized = value?.trim().toLowerCase()
+  if (!normalized) return false
+  return NEARAI_MODEL_PREFIXES.some(prefix => normalized.startsWith(prefix))
+}
+
+function applyNearaiEnvOnlyDefaults(): void {
+  const baseUrlOverride = getNearaiBaseUrlOverride()
+  const hasNearaiBaseOverride = baseUrlOverride !== undefined
+  const modelOverride = process.env.OPENAI_MODEL?.trim() || undefined
+
+  process.env.CLAUDE_CODE_USE_OPENAI = '1'
+  process.env.OPENAI_BASE_URL =
+    baseUrlOverride ?? getRouteDefaultBaseUrl('nearai')
+  process.env.OPENAI_MODEL =
+    (hasNearaiBaseOverride || isNearaiModelName(modelOverride)
+      ? modelOverride
+      : undefined) ??
+    getRouteDefaultModel('nearai')
+  process.env.OPENAI_API_KEY = process.env.NEARAI_API_KEY
+  delete process.env.OPENAI_API_FORMAT
+  delete process.env.OPENAI_AUTH_HEADER
+  delete process.env.OPENAI_AUTH_SCHEME
+  delete process.env.OPENAI_AUTH_HEADER_VALUE
+}
+
 export async function getAnthropicClient({
  apiKey,
  maxRetries,
@@ -298,6 +334,8 @@ export async function getAnthropicClient({
    envOnlyProviderRouteId === 'xiaomi-mimo' && !useMiniMaxEnvOnlyProvider
  const useXaiEnvOnlyProvider =
    envOnlyProviderRouteId === 'xai' && !useMiniMaxEnvOnlyProvider
+  const useNearaiEnvOnlyProvider =
+    envOnlyProviderRouteId === 'nearai' && !useMiniMaxEnvOnlyProvider
  if (useMiniMaxEnvOnlyProvider) {
    applyMiniMaxEnvOnlyDefaults(model)
  }
@@ -307,6 +345,9 @@ export async function getAnthropicClient({
  if (useXaiEnvOnlyProvider) {
    applyXaiEnvOnlyDefaults()
  }
+  if (useNearaiEnvOnlyProvider) {
+    applyNearaiEnvOnlyDefaults()
+  }

  const shouldUseFirstPartyAuth =
    shouldUseFirstPartyAnthropicAuth(providerOverride)
@@ -383,6 +424,7 @@ export async function getAnthropicClient({
  if (
    useXiaomiMimoEnvOnlyProvider ||
    useXaiEnvOnlyProvider ||
+    useNearaiEnvOnlyProvider ||
    isEnvTruthy(process.env.CLAUDE_CODE_USE_OPENAI) ||
    isEnvTruthy(process.env.CLAUDE_CODE_USE_GITHUB) ||
    isEnvTruthy(process.env.CLAUDE_CODE_USE_GEMINI) ||
@@ -16,9 +16,11 @@
 *   5. GEMINI_API_KEY or GOOGLE_API_KEY
 *   6. MISTRAL_API_KEY
 *   7. MINIMAX_API_KEY
- *   8. XAI_API_KEY
- *   9. Local Ollama reachable (default localhost:11434)
- *  10. Local LM Studio reachable (default localhost:1234)
+ *   8. MIMO_API_KEY
+ *   9. XAI_API_KEY
+ *  10. NEARAI_API_KEY
+ *  11. Local Ollama reachable (default localhost:11434)
+ *  12. Local LM Studio reachable (default localhost:1234)
 *
 * Local-service probes are parallelized and cheap (short timeout, no
 * request body). Env scans are synchronous and run first so we don't make
@@ -43,6 +45,7 @@ export type DetectedProviderKind =
  | 'minimax'
  | 'xiaomi-mimo'
  | 'xai'
+  | 'nearai'
  | 'ollama'
  | 'lm-studio'
  | 'gitlawb-opengateway'
@@ -171,6 +174,13 @@ export function detectProviderFromEnv(
    return { kind: 'xai', source: 'XAI_API_KEY set' }
  }

+  if (envHasNonEmpty(env, 'NEARAI_API_KEY')) {
+    return {
+      kind: 'nearai',
+      source: 'NEARAI_API_KEY set',
+    }
+  }
+
  return null
 }

@@ -41,6 +41,7 @@ const PREFERRED_PROVIDER_ORDER = [
  'minimax',
  'venice',
  'atlas-cloud',
+  'nearai',
 ] as const

 function buildValidProviders(): string[] {
@@ -285,10 +286,13 @@ export function applyProviderFlag(
              : process.env.OPENAI_API_KEY !== undefined &&
                  process.env.OPENAI_API_KEY === process.env.MINIMAX_API_KEY
                ? 'minimax'
-                : process.env.OPENAI_API_KEY !== undefined &&
-                    process.env.OPENAI_API_KEY === process.env.ATLAS_CLOUD_API_KEY
-                  ? 'atlas-cloud'
                  : process.env.OPENAI_API_KEY !== undefined &&
+                      process.env.OPENAI_API_KEY === process.env.ATLAS_CLOUD_API_KEY
+                    ? 'atlas-cloud'
+                    : process.env.OPENAI_API_KEY !== undefined &&
+                        process.env.OPENAI_API_KEY === process.env.NEARAI_API_KEY
+                      ? 'nearai'
+                      : process.env.OPENAI_API_KEY !== undefined &&
                      opengatewayApiKey !== undefined &&
                      opengatewayApiKey.length > 0 &&
                      process.env.OPENAI_API_KEY === opengatewayApiKey
@@ -408,6 +412,20 @@ export function applyProviderFlag(
      }
      break

+    case 'nearai':
+      process.env.CLAUDE_CODE_USE_OPENAI = '1'
+      applyOpenAIBaseUrlDefault(provider, defaultBaseUrl)
+      if (defaultModel) {
+        process.env.OPENAI_MODEL ??= defaultModel
+      }
+      if (model) process.env.OPENAI_MODEL = model
+      if (process.env.NEARAI_API_KEY) {
+        process.env.OPENAI_API_KEY = process.env.NEARAI_API_KEY
+      } else {
+        delete process.env.OPENAI_API_KEY
+      }
+      break
+
    default:
      process.env.CLAUDE_CODE_USE_OPENAI = '1'
      applyOpenAIBaseUrlDefault(provider, defaultBaseUrl)
@@ -98,6 +98,7 @@ const PROFILE_ENV_KEYS = [
  'VENICE_API_KEY',
  'MIMO_API_KEY',
  'ATLAS_CLOUD_API_KEY',
+  'NEARAI_API_KEY',
  'OPENCODE_API_KEY',
  DEFAULT_STARTUP_PROVIDER_ENV_VAR,
 ] as const
@@ -125,6 +126,7 @@ const SECRET_ENV_KEYS = [
  'VENICE_API_KEY',
  'MIMO_API_KEY',
  'ATLAS_CLOUD_API_KEY',
+  'NEARAI_API_KEY',
  'OPENCODE_API_KEY',
 ] as const

@@ -185,6 +187,7 @@ export type ProfileEnv = {
  VENICE_API_KEY?: string
  MIMO_API_KEY?: string
  ATLAS_CLOUD_API_KEY?: string
+  NEARAI_API_KEY?: string
  OPENCODE_API_KEY?: string
 }

@@ -209,7 +212,8 @@ type SecretValueSource = Partial<
    | 'XAI_API_KEY'
    | 'VENICE_API_KEY'
    | 'MIMO_API_KEY'
-    | 'ATLAS_CLOUD_API_KEY',
+    | 'ATLAS_CLOUD_API_KEY'
+    | 'NEARAI_API_KEY',
    string | undefined
  >
 >
@@ -1629,6 +1633,7 @@ export async function buildLaunchEnv(options: {
  // unauthenticated.
  for (const dedicatedKey of [
    'ATLAS_CLOUD_API_KEY',
+    'NEARAI_API_KEY',
    'MIMO_API_KEY',
    'VENICE_API_KEY',
  ] as const) {
@@ -44,7 +44,7 @@ import {
  type ResolvedProfileRoute,
  type ProviderPreset,
 } from '../integrations/index.js'
-import { resolveEnvOnlyProviderRouteId } from '../integrations/routeMetadata.js'
+import { isNearaiBaseUrl, resolveEnvOnlyProviderRouteId } from '../integrations/routeMetadata.js'
 import { logForDebugging } from './debug.js'
 import {
  sanitizeProfileCustomHeaders,
@@ -553,6 +553,10 @@ function isProcessEnvAlignedWithProfile(
    (profile.baseUrl?.toLowerCase().includes('atlascloud')
      ? !includeApiKey ||
        sameOptionalEnvValue(processEnv.ATLAS_CLOUD_API_KEY, profile.apiKey)
+      : true) &&
+    (isNearaiBaseUrl(profile.baseUrl)
+      ? !includeApiKey ||
+        sameOptionalEnvValue(processEnv.NEARAI_API_KEY, profile.apiKey)
      : true)
  )
 }
@@ -691,6 +695,9 @@ export function applyProviderProfileToProcessEnv(profile: ProviderProfile): void
      if (route.routeId === 'atlas-cloud' || profile.baseUrl.toLowerCase().includes('atlascloud')) {
        openAIProfileEnv.ATLAS_CLOUD_API_KEY = profile.apiKey
      }
+      if (route.routeId === 'nearai' || isNearaiBaseUrl(profile.baseUrl)) {
+        openAIProfileEnv.NEARAI_API_KEY = profile.apiKey
+      }
    }
    if (route.gatewayId === 'nvidia-nim') {
      openAIProfileEnv.NVIDIA_NIM = '1'
@@ -936,6 +943,9 @@ function buildOpenAICompatibleStartupEnv(
      if (activeProfile.baseUrl?.toLowerCase().includes('atlascloud')) {
        strictEnv.ATLAS_CLOUD_API_KEY = activeProfile.apiKey
      }
+      if (isNearaiBaseUrl(activeProfile.baseUrl)) {
+        strictEnv.NEARAI_API_KEY = activeProfile.apiKey
+      }
      return applySupportedProfileCustomHeaders(activeProfile, strictEnv)
    }
  }
@@ -976,6 +986,9 @@ function buildOpenAICompatibleStartupEnv(
    if (activeProfile.baseUrl?.toLowerCase().includes('atlascloud')) {
      env.ATLAS_CLOUD_API_KEY = activeProfile.apiKey
    }
+    if (isNearaiBaseUrl(activeProfile.baseUrl)) {
+      env.NEARAI_API_KEY = activeProfile.apiKey
+    }
  } else {
    delete env.OPENAI_API_KEY
  }
@@ -1129,6 +1142,11 @@ function buildStartupProfileFromActiveProfile(
          : null
      }

+      if (route.vendorId === 'nearai') {
+        const env = buildOpenAICompatibleStartupEnv(activeProfile)
+        return env ? { profile: 'openai', env } : null
+      }
+
      // xAI OAuth profile (provider=xai with no API key). Tag the startup
      // file with profile='xai' + XAI_CREDENTIAL_SOURCE=oauth so:
      //   1. validation accepts it at startup (no spurious
@@ -39,6 +39,7 @@ const ENV_KEYS = [
  'GOOGLE_APPLICATION_CREDENTIALS',
  'XAI_API_KEY',
  'XAI_CREDENTIAL_SOURCE',
+  'NEARAI_API_KEY',
 ] as const

 const originalEnv: Record<string, string | undefined> = {}
@@ -318,6 +319,24 @@ test('xiaomi mimo validation accepts MIMO_API_KEY without OPENAI_API_KEY', async
  await expect(getProviderValidationError(process.env)).resolves.toBeNull()
 })

+test('nearai validation accepts NEARAI_API_KEY for cloud-api base URL', async () => {
+  process.env.CLAUDE_CODE_USE_OPENAI = '1'
+  process.env.OPENAI_BASE_URL = 'https://cloud-api.near.ai/v1'
+  process.env.NEARAI_API_KEY = 'nearai-live-key'
+  delete process.env.OPENAI_API_KEY
+
+  await expect(getProviderValidationError(process.env)).resolves.toBeNull()
+})
+
+test('nearai validation accepts NEARAI_API_KEY for wildcard TEE completions endpoint', async () => {
+  process.env.CLAUDE_CODE_USE_OPENAI = '1'
+  process.env.OPENAI_BASE_URL = 'https://qwen35-122b.completions.near.ai/v1'
+  process.env.NEARAI_API_KEY = 'nearai-tee-key'
+  delete process.env.OPENAI_API_KEY
+
+  await expect(getProviderValidationError(process.env)).resolves.toBeNull()
+})
+
 test('opengateway validation fails without OPENGATEWAY_API_KEY or OPENAI_API_KEY', async () => {
  process.env.CLAUDE_CODE_USE_OPENAI = '1'
  process.env.OPENAI_BASE_URL = 'https://opengateway.gitlawb.com/v1'
@@ -15,6 +15,7 @@ import {
  getRouteCredentialValue,
  getRouteDescriptor,
  getRouteDefaultModel,
+  matchHostnameAgainstRouteHosts,
  resolveActiveRouteIdFromEnv,
  resolveRouteIdFromBaseUrl,
 } from '../integrations/routeMetadata.js'
@@ -234,9 +235,8 @@ function getRuntimeValidationTarget(
    }

    return (
-      routing.matchBaseUrlHosts?.some(
-        host => requestHost === host.toLowerCase(),
-      ) ?? false
+      routing.matchBaseUrlHosts &&
+      matchHostnameAgainstRouteHosts(requestHost, routing.matchBaseUrlHosts)
    )
  })