Files
slate/app/lib/modules/llm/base-provider.ts
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

183 lines
5.2 KiB
TypeScript

import type { LanguageModelV1 } from 'ai';
import type { ProviderInfo, ProviderConfig, ModelInfo } from './types';
import type { IProviderSetting } from '~/types/model';
import { createOpenAI } from '@ai-sdk/openai';
import { LLMManager } from './manager';
/** Default timeout for model listing API calls (5 seconds) */
const MODEL_FETCH_TIMEOUT = 5_000;
export abstract class BaseProvider implements ProviderInfo {
abstract name: string;
abstract staticModels: ModelInfo[];
abstract config: ProviderConfig;
cachedDynamicModels?: {
cacheId: string;
models: ModelInfo[];
};
getApiKeyLink?: string;
labelForGetApiKey?: string;
icon?: string;
/**
* Convert Cloudflare Env bindings to a plain Record<string, string>.
* Useful because provider methods expect Record<string, string> but
* Cloudflare Workers pass an Env interface.
*/
protected convertEnvToRecord(env?: Env): Record<string, string> {
if (!env) {
return {};
}
return Object.entries(env).reduce(
(acc, [key, value]) => {
acc[key] = String(value);
return acc;
},
{} as Record<string, string>,
);
}
/**
* Rewrite localhost / 127.0.0.1 URLs to host.docker.internal when
* running inside Docker. Only applies on the server side.
*/
protected resolveDockerUrl(baseUrl: string, serverEnv?: Record<string, string>): string {
const isDocker = process?.env?.RUNNING_IN_DOCKER === 'true' || serverEnv?.RUNNING_IN_DOCKER === 'true';
if (!isDocker) {
return baseUrl;
}
return baseUrl.replace('localhost', 'host.docker.internal').replace('127.0.0.1', 'host.docker.internal');
}
/**
* Create an AbortSignal that times out after the given milliseconds.
* Used to prevent model-listing fetches from hanging indefinitely.
*/
protected createTimeoutSignal(ms: number = MODEL_FETCH_TIMEOUT): AbortSignal {
return AbortSignal.timeout(ms);
}
getProviderBaseUrlAndKey(options: {
apiKeys?: Record<string, string>;
providerSettings?: IProviderSetting;
serverEnv?: Record<string, string>;
defaultBaseUrlKey: string;
defaultApiTokenKey: string;
}) {
const { apiKeys, providerSettings, serverEnv, defaultBaseUrlKey, defaultApiTokenKey } = options;
let settingsBaseUrl = providerSettings?.baseUrl;
const manager = LLMManager.getInstance();
if (settingsBaseUrl && settingsBaseUrl.length == 0) {
settingsBaseUrl = undefined;
}
const baseUrlKey = this.config.baseUrlKey || defaultBaseUrlKey;
let baseUrl =
settingsBaseUrl ||
serverEnv?.[baseUrlKey] ||
process?.env?.[baseUrlKey] ||
manager.env?.[baseUrlKey] ||
this.config.baseUrl;
if (baseUrl && baseUrl.endsWith('/')) {
baseUrl = baseUrl.slice(0, -1);
}
const apiTokenKey = this.config.apiTokenKey || defaultApiTokenKey;
const apiKey =
apiKeys?.[this.name] || serverEnv?.[apiTokenKey] || process?.env?.[apiTokenKey] || manager.env?.[apiTokenKey];
return {
baseUrl,
apiKey,
};
}
getModelsFromCache(options: {
apiKeys?: Record<string, string>;
providerSettings?: Record<string, IProviderSetting>;
serverEnv?: Record<string, string>;
}): ModelInfo[] | null {
if (!this.cachedDynamicModels) {
return null;
}
const cacheKey = this.cachedDynamicModels.cacheId;
const generatedCacheKey = this.getDynamicModelsCacheKey(options);
if (cacheKey !== generatedCacheKey) {
this.cachedDynamicModels = undefined;
return null;
}
return this.cachedDynamicModels.models;
}
getDynamicModelsCacheKey(options: {
apiKeys?: Record<string, string>;
providerSettings?: Record<string, IProviderSetting>;
serverEnv?: Record<string, string>;
}) {
// Only include provider-relevant env keys, not the entire server environment
const relevantEnvKeys = [this.config.baseUrlKey, this.config.apiTokenKey].filter(Boolean) as string[];
const relevantEnv: Record<string, string> = {};
for (const key of relevantEnvKeys) {
if (options.serverEnv?.[key]) {
relevantEnv[key] = options.serverEnv[key];
}
}
return JSON.stringify({
apiKeys: options.apiKeys?.[this.name],
providerSettings: options.providerSettings?.[this.name],
serverEnv: relevantEnv,
});
}
storeDynamicModels(
options: {
apiKeys?: Record<string, string>;
providerSettings?: Record<string, IProviderSetting>;
serverEnv?: Record<string, string>;
},
models: ModelInfo[],
) {
const cacheId = this.getDynamicModelsCacheKey(options);
this.cachedDynamicModels = {
cacheId,
models,
};
}
// Declare the optional getDynamicModels method
getDynamicModels?(
apiKeys?: Record<string, string>,
settings?: IProviderSetting,
serverEnv?: Record<string, string>,
): Promise<ModelInfo[]>;
abstract getModelInstance(options: {
model: string;
serverEnv?: Env;
apiKeys?: Record<string, string>;
providerSettings?: Record<string, IProviderSetting>;
}): LanguageModelV1;
}
type OptionalApiKey = string | undefined;
export function getOpenAILikeModel(baseURL: string, apiKey: OptionalApiKey, model: string) {
const openai = createOpenAI({
baseURL,
apiKey,
});
return openai(model);
}