Packaging

Embeddings priced by parameter size

Fireworks AI pricing

Fireworks introduces parameter-size-based pricing for its embeddings models.

From Fireworks AI's pricing timeline
Embeddings Pricing by Parameter Size

Fireworks published differential embeddings pricing by base model parameter count: <150M params at $0.008/1M tokens, 150–350M at $0.016/1M, Qwen3 8B at $0.10/1M. The schedule undercuts OpenAI text-embedding-3-small ($0.02/1M) by 60% for the smallest tier and creates a granular cost ladder for retrieval-pipeline cost optimization.

About Fireworks AI
fireworks.ai ↗

Fireworks AI runs a pure-usage inference platform: serverless per-token endpoints for popular open-weight models (Llama, DeepSeek, Mixtral, Qwen) plus on-demand dedicated GPU deployments at $7/hr H100/H200, $10/hr B200, $12/hr B300.

Free tier
Yes
Commits
Available
Transparency
public

Fireworks AI pricing history

  1. Feb 2026
    Embeddings Pricing by Parameter Size
  2. Aug 2025
    B200 and B300 GPU Availability — Frontier Pricing
  3. Mar 2025
    Batch API at 50% Discount Across All Models
  4. Nov 2024
    Turbo + Priority Tiers + Cached Input Discount
  5. Jul 2024
    Series B ($52M) at $552M Valuation
Full Fireworks AI timeline
All pricing activity