New Provider: ElevenLabs Speech Models and 12 New Cloudflare AI Models Added

Date: 2026-02-03 / (1404-11-14)

Summary

AvalAI adds ElevenLabs as a new provider with 7 speech models offering advanced text-to-speech and speech-to-text capabilities. Additionally, 12 new Cloudflare AI models are now available, including FLUX image generation, embedding models, and chat completion models from multiple providers.

---

Details

ElevenLabs - New Speech Provider

ElevenLabs is now available on AvalAI, bringing industry-leading text-to-speech and speech-to-text capabilities. ElevenLabs models deliver natural, expressive speech synthesis and accurate transcription across multiple languages. Documentation

Text-to-Speech Models

Five TTS models are now available via the v1/audio/speech endpoint:

Model	Description	Latency	Languages
eleven_turbo_v2	High quality, low-latency English TTS	~250-300ms	English
eleven_turbo_v2_5	High quality, low-latency multilingual TTS	~250-300ms	32 languages
eleven_flash_v2	Ultra-fast English TTS	~75ms	English
eleven_flash_v2_5	Ultra-fast multilingual TTS	~75ms	32 languages
eleven_multilingual_v2	Most lifelike, emotionally rich TTS	Higher	29 languages

Key Features:

• Natural Speech: Human-like and expressive voice generation
• Multi-language Support: Up to 32 languages with Flash v2.5 and Turbo v2.5
• Low Latency: Ultra-fast ~75ms latency with Flash models
• OpenAI-Compatible: Standard v1/audio/speech endpoint

Pricing:

Model	Output Cost (per second)
eleven_turbo_v2	$0.0025/sec
eleven_turbo_v2_5	$0.0025/sec
eleven_flash_v2	$0.0025/sec
eleven_flash_v2_5	$0.0025/sec
eleven_multilingual_v2	$0.005/sec

Speech-to-Text Models

Two state-of-the-art transcription models via the v1/audio/transcriptions endpoint:

Model	Description	Languages
scribe_v2	State-of-the-art speech recognition	90+ languages
scribe_v1	Legacy transcription model	90+ languages

Scribe v2 Features:

• Accurate Transcription: 90+ languages supported
• Keyterm Prompting: Up to 100 terms for domain-specific accuracy
• Entity Detection: Up to 56 entity types
• Word-level Timestamps: Precise timing for each word
• Speaker Diarization: Up to 32 speakers identification
• Smart Language Detection: Automatic language identification

Pricing:

Model	Input Cost (per second)
scribe_v2	$0.00009722/sec
scribe_v1	$0.00009722/sec

---

Cloudflare AI - 12 New Models

We've expanded our Cloudflare AI offerings with 12 new models spanning image generation, embeddings, and chat completions. Documentation

Image Generation Models (v1/images/generations)

Five new image generation models are now available:

Model	Description	Base Price (1MP)
cf.flux-2-klein-9b	FLUX 2 Klein 9B parameter model	$0.015/image
cf.flux-2-klein-4b	FLUX 2 Klein 4B compact model	$0.010/image
cf.flux-2-dev	FLUX 2 Development model	$0.010/image
cf.lucid-origin	Lucid Origin image generation	$0.015/image
cf.phoenix-1.0	Phoenix 1.0 image generation	$0.015/image

Resolution-based Pricing:

Model	1MP	2MP	3MP	4MP
cf.flux-2-klein-9b	$0.015	$0.017	$0.019	$0.021
cf.flux-2-klein-4b	$0.010	$0.012	$0.014	$0.016
cf.flux-2-dev	$0.010	$0.011	$0.012	$0.013
cf.lucid-origin	$0.015	$0.017	$0.019	$0.021
cf.phoenix-1.0	$0.015	$0.017	$0.019	$0.021

Embedding Models (v1/embeddings)

Three new embedding models for semantic search and text analysis:

Model	Description	Input Price ($/1M tokens)
cf.qwen3-embedding-0.6b	Qwen3 Embedding 0.6B from Alibaba	$0.012
cf.plamo-embedding-1b	PLaMo Embedding 1B from PFN	$0.019
cf.embeddinggemma-300m	EmbeddingGemma 300M from Google	$0.012

Chat Completion Models (v1/chat/completions, v1/responses)

Four new chat models with partial v1/responses support:

Model	Owner	Input ($/1M)	Cached Input ($/1M)	Output ($/1M)
cf.qwen3-30b-a3b-fp8	Alibaba	$0.051	$0.025	$0.34
cf.granite-4.0-h-micro	IBM	$0.017	$0.008	$0.11
cf.gpt-oss-120b	OpenAI	$0.35	$0.175	$0.75
cf.gpt-oss-20b	OpenAI	$0.20	$0.10	$0.30

---

API Request/Response Examples

ElevenLabs Text-to-Speech Example

curl https://api.avalai.ir/v1/audio/speech \
  -H "Authorization: Bearer $AVALAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "eleven_flash_v2_5",
    "input": "The quick brown fox jumped over the lazy dog.",
    "voice": "coral"
  }' \
  --output speech.mp3

ElevenLabs Transcription Example

curl https://api.avalai.ir/v1/audio/transcriptions \
  -H "Authorization: Bearer $AVALAI_API_KEY" \
  -F file=@audio.mp3 \
  -F model=scribe_v2

Example Response:

{
  "text": "The quick brown fox jumped over the lazy dog.",
  "task": "transcribe",
  "language": "en",
  "duration": 2.5,
  "words": [
    {"word": "The", "start": 0.0, "end": 0.15},
    {"word": "quick", "start": 0.15, "end": 0.35},
    ...
  ]
}

Cloudflare Image Generation Example

curl https://api.avalai.ir/v1/images/generations \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $AVALAI_API_KEY" \
  -d '{
    "model": "cf.flux-2-klein-9b",
    "prompt": "A futuristic cityscape at sunset with flying cars",
    "n": 1,
    "size": "1024x1024"
  }'

Cloudflare Embedding Example

curl https://api.avalai.ir/v1/embeddings \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $AVALAI_API_KEY" \
  -d '{
    "model": "cf.qwen3-embedding-0.6b",
    "input": "Machine learning enables computers to learn from data."
  }'

---

Price Comparison

ElevenLabs TTS Models

Model	Output Cost/Second	Best For
eleven_flash_v2_5	$0.0025	Real-time apps, low latency
eleven_turbo_v2_5	$0.0025	Balanced quality and speed
eleven_multilingual_v2	$0.005	Highest quality, emotional range

Cloudflare Image Models

Model	Per Image (1MP)	Best For
cf.flux-2-klein-4b	$0.010	Cost-effective generation
cf.flux-2-dev	$0.010	Development and testing
cf.flux-2-klein-9b	$0.015	Higher quality output

---

Related Resources

• ElevenLabs Models Documentation
• Cloudflare Models Documentation
• Audio API Reference
• Images API Reference
• Embeddings API Reference
• Official ElevenLabs API Documentation