voyage-finance-2
Finance-domain embedding tuned for filings, earnings, and financial QA, with a 32K context window.
About
voyage-finance-2 is Voyage AI's finance-specialist embedding, launched in June 2024 to complete the domain trio with law-2 and code-2. It is trained to handle financial jargon and the structure of SEC filings, earnings reports, financial news, and hybrid tabular/text question answering, which generic embeddings handle poorly.
Voyage's launch post claims a 7% average gain over OpenAI and 12% over Cohere on financial retrieval datasets, and the model supports a 32K context length that was unusually large for its class at launch. Like its sibling it costs $0.12 per million tokens with a 50-million-token free tier, served through MongoDB's Voyage AI endpoints after the February 2025 acquisition.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
- 7% average retrieval gain over OpenAI (next best) and 12% over Cohere on financial retrieval datasets (provider-reported, launch blog)
- Evaluated across 11 financial datasets incl. SEC filings and earnings reports (provider-reported)
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 8/10
Scored 8/10 (verified tier, metrics-v1). Evidence: 32K context, 1024 dims, $0.12/1M pricing, and 50M free tokens confirmed on MongoDB's official docs; June 2024 launch date from Voyage's own blog. Strengths: the only serious finance-tuned embedding API, long context, stable provider post-acquisition. Docked for: (1) weights closed; (2) the 7%/12% gain figures are provider-run on provider-chosen datasets.
How the trust score is built
Full methodology →- Adoption · stars/downloads, log-scaled vs peers
- 8.0/10
- Maintenance · recency, releases, issue load
- 8.0/10
- Openness · license + docs
- 3.0/10
- Manual review · sanity read + red-flag check
- 8.0/10
- · Launched 2024-06-03 per Voyage's own blog; 32K context, 1024 dims, $0.12/1M, 50M free tokens
- · Provider claims 7% avg gain over OpenAI, 12% over Cohere on financial retrieval datasets (provider-chosen evals)
- · Only serious finance-tuned embedding API on the market