Files
Stijnus b7ef2247b8 feat: add Cerebras and Fireworks AI LLM providers (#2113)
* fix: improve local model provider robustness and UX

- Extract shared Docker URL rewriting and env conversion into BaseProvider
  to eliminate 4x duplicated code across Ollama and LMStudio
- Add error handling and 5s timeouts to all model-listing fetches so one
  unreachable provider doesn't block the entire model list
- Fix Ollama using createOllama() instead of mutating provider internals
- Fix LLMManager singleton ignoring env updates on subsequent requests
- Narrow cache key to only include provider-relevant env vars instead of
  the entire server environment
- Fix 'as any' casts in LMStudio and OpenAILike by using shared
  convertEnvToRecord helper
- Replace console.log/error with structured logger in OpenAILike
- Fix typo: filteredStaticModesl -> filteredStaticModels in manager
- Add connection status indicator (green/red dot) for local providers
  in the ModelSelector dropdown
- Show helpful "is X running?" message when local provider has no models

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* feat: add Cerebras LLM provider

- Add Cerebras provider with 8 models (Llama, GPT OSS, Qwen, ZAI GLM)
- Integrate @ai-sdk/cerebras@0.2.16 for compatibility
- Add CEREBRAS_API_KEY to environment configuration
- Register provider in LLMManager registry

Models included:
- llama3.1-8b, llama-3.3-70b
- gpt-oss-120b (reasoning)
- qwen-3-32b, qwen-3-235b variants
- zai-glm-4.6, zai-glm-4.7 (reasoning)

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

* feat: add Fireworks AI LLM provider

- Add Fireworks provider with 6 popular models
- Integrate @ai-sdk/fireworks@0.2.16 for compatibility
- Add FIREWORKS_API_KEY to environment configuration
- Register provider in LLMManager registry

Models included:
- Llama 3.1 variants (405B, 70B, 8B Instruct)
- DeepSeek R1 (reasoning model)
- Qwen 2.5 72B Instruct
- FireFunction V2

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

* feat: add coding-specific models to existing providers

Enhanced providers with state-of-the-art coding models:

**DeepSeek Provider:**
+ DeepSeek V3.2 (integrates thinking + tool-use)
+ DeepSeek V3.2-Speciale (high-compute variant, beats GPT-5)

**Fireworks Provider:**
+ Qwen3-Coder 480B (262K context, best for coding)
+ Qwen3-Coder 30B (fast coding specialist)

**Cerebras Provider:**
+ Qwen3-Coder 480B (2000 tokens/sec!)
- Removed deprecated models (qwen-3-32b, llama-3.3-70b)

Total new models: 4
Total coding models across all providers: 12+

Performance highlights:
- Qwen3-Coder: State-of-the-art coding performance
- DeepSeek V3.2: Integrates thinking directly into tool-use
- ZAI GLM 4.6: 73.8% SWE-bench score
- Ultra-fast inference: 2000 tok/s on Cerebras

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

* feat: add dynamic model discovery to providers

Implemented getDynamicModels() for automatic model discovery:

**DeepSeek Provider:**
- Fetches models from https://api.deepseek.com/models
- Automatically discovers new models as DeepSeek adds them
- Filters out static models to avoid duplicates

**Cerebras Provider:**
- Fetches models from https://api.cerebras.ai/v1/models
- Auto-discovers new Cerebras models
- Keeps UI up-to-date with latest offerings

**Fireworks Provider:**
- Fetches from https://api.fireworks.ai/v1/accounts/fireworks/models
- Includes context_length from API response
- Discovers new Qwen-Coder and other models automatically

**Moonshot Provider:**
- Fetches from https://api.moonshot.ai/v1/models
- OpenAI-compatible endpoint
- Auto-discovers new Kimi models

Benefits:
-  No manual updates needed when providers add new models
-  Users always have access to latest models
-  Graceful fallback to static models if API fails
-  5-second timeout prevents hanging
-  Caching system built into BaseProvider

Technical details:
- Uses BaseProvider's built-in caching system
- Cache invalidates when API keys change
- Failed API calls fallback to static models
- All endpoints have 5-second timeout protection

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

* feat: add Z.AI provider with GLM models and JWT authentication

Merged changes from PR #2069 to add Z.AI provider:
- Added GLM-4.6 (200K), GLM-4.5 (128K), and GLM-4.5 Flash models
- Implemented secure JWT token generation with HMAC-SHA256 signing
- Added dynamic model discovery from Z.AI API
- Included proper error handling and token validation
- GLM-4.6 achieves 73.8% on SWE-bench coding benchmarks

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-07 15:26:46 +01:00

194 lines
5.5 KiB
TypeScript

import { BaseProvider } from '~/lib/modules/llm/base-provider';
import type { IProviderSetting } from '~/types/model';
import type { LanguageModelV1 } from 'ai';
import type { ModelInfo } from '~/lib/modules/llm/types';
import { createOpenAI } from '@ai-sdk/openai';
import crypto from 'node:crypto';
export default class ZaiProvider extends BaseProvider {
name = 'Z.ai';
getApiKeyLink = 'https://open.bigmodel.cn/usercenter/apikeys';
config = {
baseUrlKey: 'ZAI_BASE_URL',
apiTokenKey: 'ZAI_API_KEY',
baseUrl: 'https://api.z.ai/api/coding/paas/v4', //Dedicated endpoint for Coding Plan
};
staticModels: ModelInfo[] = [
{
name: 'glm-4.6',
label: 'GLM-4.6 (200K)',
provider: 'Z.ai',
maxTokenAllowed: 200000,
maxCompletionTokens: 65536,
},
{
name: 'glm-4.5',
label: 'GLM-4.5 (128K)',
provider: 'Z.ai',
maxTokenAllowed: 128000,
maxCompletionTokens: 65536,
},
{
name: 'glm-4.5-flash',
label: 'GLM-4.5 Flash (128K)',
provider: 'Z.ai',
maxTokenAllowed: 128000,
maxCompletionTokens: 65536,
},
];
async getDynamicModels(
apiKeys?: Record<string, string>,
settings?: IProviderSetting,
serverEnv?: Record<string, string>,
): Promise<ModelInfo[]> {
const { baseUrl, apiKey } = this.getProviderBaseUrlAndKey({
apiKeys,
providerSettings: settings,
serverEnv: serverEnv as any,
defaultBaseUrlKey: 'ZAI_BASE_URL',
defaultApiTokenKey: 'ZAI_API_KEY',
});
if (!apiKey) {
throw new Error(`Missing Api Key configuration for ${this.name} provider`);
}
const token = this._generateToken(apiKey);
if (!this._isValidToken(token)) {
throw new Error(`Invalid API key format for ${this.name} provider`);
}
try {
const response = await fetch(`${baseUrl}/models`, {
headers: {
Authorization: `Bearer ${token}`,
'Content-Type': 'application/json',
},
});
if (!response.ok) {
throw new Error(`Failed to fetch models: ${response.status} ${response.statusText}`);
}
const res = (await response.json()) as any;
const staticModelIds = this.staticModels.map((m) => m.name);
// Filter out static models and only include GLM models
const data =
res.data?.filter(
(model: any) =>
model.object === 'model' && model.id?.startsWith('glm-') && !staticModelIds.includes(model.id),
) || [];
return data.map((m: any) => {
let contextWindow = 128000;
let maxCompletionTokens = 65536;
if (m.id?.includes('glm-4.6')) {
contextWindow = 200000;
maxCompletionTokens = 65536;
} else if (m.id?.includes('glm-4.5')) {
contextWindow = 128000;
maxCompletionTokens = 65536;
} else if (m.id?.includes('glm-4')) {
contextWindow = 128000;
maxCompletionTokens = 8192;
} else if (m.id?.includes('glm-3')) {
contextWindow = 32000;
maxCompletionTokens = 4096;
}
return {
name: m.id,
label: `${m.id} (${Math.floor(contextWindow / 1000)}k context)`,
provider: this.name,
maxTokenAllowed: contextWindow,
maxCompletionTokens,
};
});
} catch (error) {
console.error(`Failed to fetch dynamic models for ${this.name}:`, error);
return [];
}
}
private _generateToken(apiKey: string): string {
try {
const [id, secret] = apiKey.split('.');
if (!id || !secret) {
throw new Error(`Invalid API key format for ${this.name}. Expected: id.secret`);
}
const now = Date.now();
const payload = {
apiKey: id,
exp: now + 3600 * 1000,
timestamp: now,
};
const header = { alg: 'HS256', sign_type: 'SIGN' };
const base64Url = (obj: any) =>
Buffer.from(JSON.stringify(obj)).toString('base64').replace(/=/g, '').replace(/\+/g, '-').replace(/\//g, '_');
const signature = crypto
.createHmac('sha256', secret)
.update(base64Url(header) + '.' + base64Url(payload))
.digest('base64')
.replace(/=/g, '')
.replace(/\+/g, '-')
.replace(/\//g, '_');
return `${base64Url(header)}.${base64Url(payload)}.${signature}`;
} catch (error) {
console.error(`Failed to generate JWT token for ${this.name}:`, error);
throw new Error(`Failed to generate JWT token: ${error instanceof Error ? error.message : 'Unknown error'}`);
}
}
/**
* Validates JWT token format
*/
private _isValidToken(token: string): boolean {
try {
const parts = token.split('.');
return parts.length === 3 && parts.every((part) => part.length > 0);
} catch {
return false;
}
}
getModelInstance(options: {
model: string;
serverEnv: Env;
apiKeys?: Record<string, string>;
providerSettings?: Record<string, IProviderSetting>;
}): LanguageModelV1 {
const { model, serverEnv, apiKeys, providerSettings } = options;
const { baseUrl, apiKey } = this.getProviderBaseUrlAndKey({
apiKeys,
providerSettings: providerSettings?.[this.name],
serverEnv: serverEnv as any,
defaultBaseUrlKey: 'ZAI_BASE_URL',
defaultApiTokenKey: 'ZAI_API_KEY',
});
if (!apiKey) {
throw new Error(`Missing API key for ${this.name} provider`);
}
const token = this._generateToken(apiKey);
const zaiClient = createOpenAI({
baseURL: baseUrl,
apiKey: token,
});
return zaiClient(model);
}
}