Files
bolt-diy/app/lib/modules/llm/providers/anthropic.ts
Stijnus b5d9055851 🔧 Fix Token Limits & Invalid JSON Response Errors (#1934)
ISSUES FIXED:
-  Invalid JSON response errors during streaming
-  Incorrect token limits causing API rejections
-  Outdated hardcoded model configurations
-  Poor error messages for API failures

SOLUTIONS IMPLEMENTED:

🎯 ACCURATE TOKEN LIMITS & CONTEXT SIZES
- OpenAI GPT-4o: 128k context (was 8k)
- OpenAI GPT-3.5-turbo: 16k context (was 8k)
- Anthropic Claude 3.5 Sonnet: 200k context (was 8k)
- Anthropic Claude 3 Haiku: 200k context (was 8k)
- Google Gemini 1.5 Pro: 2M context (was 8k)
- Google Gemini 1.5 Flash: 1M context (was 8k)
- Groq Llama models: 128k context (was 8k)
- Together models: Updated with accurate limits

�� DYNAMIC MODEL FETCHING ENHANCED
- Smart context detection from provider APIs
- Automatic fallback to known limits when API unavailable
- Safety caps to prevent token overflow (100k max)
- Intelligent model filtering and deduplication

🛡️ IMPROVED ERROR HANDLING
- Specific error messages for Invalid JSON responses
- Token limit exceeded warnings with solutions
- API key validation with clear guidance
- Rate limiting detection and user guidance
- Network timeout handling

 PERFORMANCE OPTIMIZATIONS
- Reduced static models from 40+ to 12 essential
- Enhanced streaming error detection
- Better API response validation
- Improved context window display (shows M/k units)

🔧 TECHNICAL IMPROVEMENTS
- Dynamic model context detection from APIs
- Enhanced streaming reliability
- Better token limit enforcement
- Comprehensive error categorization
- Smart model validation before API calls

IMPACT:
 Eliminates Invalid JSON response errors
 Prevents token limit API rejections
 Provides accurate model capabilities
 Improves user experience with clear errors
 Enables full utilization of modern LLM context windows
2025-08-29 20:53:57 +02:00

113 lines
3.5 KiB
TypeScript

import { BaseProvider } from '~/lib/modules/llm/base-provider';
import type { ModelInfo } from '~/lib/modules/llm/types';
import type { LanguageModelV1 } from 'ai';
import type { IProviderSetting } from '~/types/model';
import { createAnthropic } from '@ai-sdk/anthropic';
export default class AnthropicProvider extends BaseProvider {
name = 'Anthropic';
getApiKeyLink = 'https://console.anthropic.com/settings/keys';
config = {
apiTokenKey: 'ANTHROPIC_API_KEY',
};
staticModels: ModelInfo[] = [
/*
* Essential fallback models - only the most stable/reliable ones
* Claude 3.5 Sonnet: 200k context, excellent for complex reasoning and coding
*/
{
name: 'claude-3-5-sonnet-20241022',
label: 'Claude 3.5 Sonnet',
provider: 'Anthropic',
maxTokenAllowed: 200000,
},
// Claude 3 Haiku: 200k context, fastest and most cost-effective
{
name: 'claude-3-haiku-20240307',
label: 'Claude 3 Haiku',
provider: 'Anthropic',
maxTokenAllowed: 200000,
},
];
async getDynamicModels(
apiKeys?: Record<string, string>,
settings?: IProviderSetting,
serverEnv?: Record<string, string>,
): Promise<ModelInfo[]> {
const { apiKey } = this.getProviderBaseUrlAndKey({
apiKeys,
providerSettings: settings,
serverEnv: serverEnv as any,
defaultBaseUrlKey: '',
defaultApiTokenKey: 'ANTHROPIC_API_KEY',
});
if (!apiKey) {
throw `Missing Api Key configuration for ${this.name} provider`;
}
const response = await fetch(`https://api.anthropic.com/v1/models`, {
headers: {
'x-api-key': `${apiKey}`,
'anthropic-version': '2023-06-01',
},
});
const res = (await response.json()) as any;
const staticModelIds = this.staticModels.map((m) => m.name);
const data = res.data.filter((model: any) => model.type === 'model' && !staticModelIds.includes(model.id));
return data.map((m: any) => {
// Get accurate context window from Anthropic API
let contextWindow = 32000; // default fallback
// Anthropic provides max_tokens in their API response
if (m.max_tokens) {
contextWindow = m.max_tokens;
} else if (m.id?.includes('claude-3-5-sonnet')) {
contextWindow = 200000; // Claude 3.5 Sonnet has 200k context
} else if (m.id?.includes('claude-3-haiku')) {
contextWindow = 200000; // Claude 3 Haiku has 200k context
} else if (m.id?.includes('claude-3-opus')) {
contextWindow = 200000; // Claude 3 Opus has 200k context
} else if (m.id?.includes('claude-3-sonnet')) {
contextWindow = 200000; // Claude 3 Sonnet has 200k context
}
return {
name: m.id,
label: `${m.display_name} (${Math.floor(contextWindow / 1000)}k context)`,
provider: this.name,
maxTokenAllowed: contextWindow,
};
});
}
getModelInstance: (options: {
model: string;
serverEnv: Env;
apiKeys?: Record<string, string>;
providerSettings?: Record<string, IProviderSetting>;
}) => LanguageModelV1 = (options) => {
const { apiKeys, providerSettings, serverEnv, model } = options;
const { apiKey } = this.getProviderBaseUrlAndKey({
apiKeys,
providerSettings,
serverEnv: serverEnv as any,
defaultBaseUrlKey: '',
defaultApiTokenKey: 'ANTHROPIC_API_KEY',
});
const anthropic = createAnthropic({
apiKey,
headers: { 'anthropic-beta': 'output-128k-2025-02-19' },
});
return anthropic(model);
};
}