Calculator
Calculateur Coût Embeddings IA
Estimez le coût d'embedding unique et récurrent entre 9+ providers. Branchez taille du corpus, stratégie de chunks, fréquence de refresh.
Pricing refreshed: 2026-05-01
Cheapest · year 1
Together · BGE-M3
1024 dim · 8,192 max tokens
$2
| Provider | Model | $ / 1M tokens | Coût embed unique | Coût mensuel | Year 1 |
|---|---|---|---|---|---|
| Together | BGE-M3 1024 dim · Self-host open weights for $0 | $0.008 | $0.40 | $0.14 | $2 |
| Together | bge-large-en-v1.5 1024 dim | $0.008 | $0.40 | $0.14 | $2 |
| Fireworks | nomic-embed-text-v1.5 768 dim | $0.008 | $0.40 | $0.14 | $2 |
| Jina AI | jina-embeddings-v3 1024 dim · configurable | $0.012 | $0.60 | $0.21 | $3 |
| Jina AI | jina-embeddings-v4 2048 dim · configurable | $0.018 | $0.90 | $0.31 | $5 |
| OpenAI | text-embedding-3-small 1536 dim · configurable | $0.02 | $1.00 | $0.35 | $5 |
| Voyage AI | voyage-3-lite 512 dim | $0.02 | $1.00 | $0.35 | $5 |
| Amazon Bedrock | Titan Embed v2 1024 dim · configurable | $0.02 | $1.00 | $0.35 | $5 |
| text-embedding-005 768 dim | $0.025 | $1.25 | $0.44 | $7 | |
| Voyage AI | voyage-3 1024 dim | $0.06 | $3.00 | $1.05 | $16 |
| Cohere | embed-english-v3.0 1024 dim | $0.10 | $5.00 | $1.75 | $26 |
| Cohere | embed-multilingual-v3.0 1024 dim | $0.10 | $5.00 | $1.75 | $26 |
| Cohere | embed-english-light-v3.0 384 dim · Smaller, cheaper at inference | $0.10 | $5.00 | $1.75 | $26 |
| gemini-embedding-exp 3072 dim · configurable | $0.10 | $5.00 | $1.75 | $26 | |
| Mistral | mistral-embed 1024 dim | $0.10 | $5.00 | $1.75 | $26 |
| OpenAI | text-embedding-3-large 3072 dim · configurable · Matryoshka — truncate to 256/512/1024 without retrain | $0.13 | $6.50 | $2.28 | $34 |
| Voyage AI | voyage-3-large 1024 dim · configurable · Top MTEB benchmarks 2025-2026 | $0.18 | $9.00 | $3.15 | $47 |
| Voyage AI | voyage-code-3 1024 dim · Optimized for code retrieval | $0.18 | $9.00 | $3.15 | $47 |
Refresh frequency of 0.25 means re-embed the corpus once every 4 months. Models marked "configurable" support Matryoshka truncation — you can downsize dimensions post-hoc without re-embedding.