Replicate vs Together AI
A side-by-side look at pricing, capabilities, pros, cons, and our editorial scores.
Tagline
Replicate
One-API platform for running and fine-tuning open-source models.Together AI
Fine-tune & serve open-weight models (Llama, Mistral, DeepSeek).Pricing
Replicate
PaidΒ· Pay-per-second of GPU timeTogether AI
PaidΒ· Pay-per-token; fine-tuning per-tokenPlatforms
Replicate
api
Together AI
api
Open source
Replicate
Not listed Β· Apache-2.0Together AI
Not listedGitHub stars
Replicate
597
checked 2026-11-13
Together AI
βLast GitHub push
Replicate
2025-11-17Together AI
βFirst commit
Replicate
2022-11Together AI
βPackage downloads (30 d)
Replicate
3,263,078 Β· npm replicate
checked 2026-11-20
Together AI
βHacker News stories
Replicate
110
checked 2026-11-20
Together AI
βHeadquarters
Replicate
βTogether AI
United States
Wikidata
Company
Replicate
βTogether AI
Together AIModel used
Replicate
Thousands of community + first-party modelsTogether AI
Llama / Mistral / Qwen / DeepSeek and othersBest for
Replicate
Pick Replicate to experiment across many models or to fine-tune popular open models with minimal setup.Together AI
Pick Together when you want open-weight FT + serving in one platform with sensible per-token pricing.Not for
Replicate
Skip it for high-volume production inference where dedicated deployments or self-hosting are cheaper.Together AI
Skip it if you need the polish of OpenAI's developer experience or single-vendor support across closed + open.Editorial score
Replicate
8.5 / 10Together AI
8.6 / 10Use cases
Replicate
model hostingfine-tuningAPI access
Together AI
open modelsfine-tuninginference
Pros
Replicate
- One API, thousands of models
- Easy fine-tuning of Llama, SD, Flux
- Strong community
- Predictable per-second pricing
Together AI
- Wide open-model catalogue
- Competitive inference pricing
- Fine-tune + serve in one place
- Dedicated endpoints for production
Cons
Replicate
- Per-second pricing can surprise
- Hosted models vary in quality
Together AI
- Latency varies by model
- Less polish than OpenAI
Editorial score: rule-based, 0β10, from AI-assisted profile inputs (see /methodology) β not a user rating; βββ means unscored. βNot listedβ means we have no record of it, not that it is absent. GitHub figures and prices carry the date they were checked or captured; prices are shown as published, unconverted.
Pick Replicate if
- β One API, thousands of models
- β Easy fine-tuning of Llama, SD, Flux
- β Strong community
- β Predictable per-second pricing
Pick Together AI if
- β Wide open-model catalogue
- β Competitive inference pricing
- β Fine-tune + serve in one place
- β Dedicated endpoints for production