Use · Together AI
Is it good enough?
Ownership levelPartialnone·limited·partial·substantial·fullAnalytical input C ยท 69.6/100
This page is a projection of the one entry record, the Reliability factor that Use covers. The full verdict is set by all four factors together, floor-weighted so the weakest caps the whole.
Which domain expands which factor
- AssessUse & modify + Transparency
- ImplementData control + Reliability
- UseReliability
- SupportTransparency
Models served
The serverless chat catalogue (24 models) is led by DeepSeek V4 Pro 0813, V4.1
Flash and V4.1 Flash 1017; GLM-5.2, 5.3 and 5.3 Flash; Kimi K3 and K3.1 Pro; MiniMax M3.1 Flash;
Qwen3.5 9B, 3.6 Plus, 3.7 Max and Plus, 3.8 Flash and Qwen4 Max; gpt-oss-120B; Meta Muse Glimmer
30B and Muse Spark Open 816B; Reflection Beam 1.0; Thinking Machines Inkling 2 Small; NVIDIA
Nemotron 3.5 Super VL; and a few smaller or free models. Vision, image, video and audio models are
also served, including wrapped third-party APIs. Not on serverless chat: Llama 3.3 70B,
Gemma 4 31B, Cogito v2.1, MiniMax M3 and M2.7, Qwen3-235B (Llama, Gemma 4 and Mixtral are
fine-tuning base models only). There are no serverless embedding, rerank or moderation models.
The list turns over quickly, so check the live table before relying on a model.
Inference features
The docs list OpenAI-compatible chat, completions, vision, image generation, text-to-speech and
speech-to-text, plus a Messages API and a Responses API. The catalogue table marks function
calling and structured outputs as supported for nearly every chat model. It also states the
quantisation of each model (FP4, FP8, NVFP4, MXFP4), which contradicts the docs claim that models
are hosted at full precision.
Context & token limits
The catalogue table publishes context per model: gpt-oss-120B 131,072; Qwen3.5 9B 262,144; DeepSeek
V4 Pro 0813 and Kimi K3 about 1.05 million; DeepSeek V4.1 Flash 1,000,000; GLM-5.3 1,048,575; Meta
Muse Glimmer 30B 131,072; Together Tev1 4B 32,768. Maximum output tokens are not listed there.
Fine-tuning / custom models
Fine-tuning and post-training are sold and priced on the pricing page. LoRA and full fine-tuning
are billed per 1M training tokens (from $0.34 for Qwen3.5 0.8B to $40 for GLM-5.1 and 5.2, with
minimum charges of $4 to $60) over Llama, Qwen, DeepSeek, Gemma, gpt-oss, Kimi and GLM base
models, and GPU-hour post-training is also offered. Methods, data formats and the hosting terms
for tuned models are not stated.
How this scores
The ownership factor this domain covers, drawn from the one entry record.
3
ReliabilityDoes it stay up and stay secure?
ModerateA public status page shows per-model 90-day uptime of 99.28% to 100%, and SOC 2 Type II and ISO 27001:2022 are announced. The terms disclaim uptime, serverless is best-effort, and the only SLA is sales-only provisioned throughput.
How this scores (AOI sub-dimensions)
Reliability3/5whether it stays up, with an SLA and status historyA public status page exists at status.together.ai (ev-status).
Security4/5the controls protecting your traffic and dataThe DPA Annex 2 (ev9) commits to a written security programme, MFA for remote access, and encryption in transit and storage.
Compliance4/5which independent certifications and attestations it holdsSOC 2 Type II (ev3) and ISO 27001:2022 from A-LIGN, an ANAB-accredited body (ev-iso), are both vendor announcements of independent audits.
What this means for adoptionYou own your inputs and outputs under the Terms, training is opt-in, and the open-weight catalogue on an OpenAI-compatible API gives a clean portable exit. Ownership is partial. By default Together stores your prompts and may use them for product improvements, and zero data retention is an admin opt-in. Passthrough models send prompts to third parties until you switch them off, and the Terms disclaim any duty of confidence. Residency is US-only on serverless, with EU limited to sales-led dedicated deployments. Turn prompt storage off for ZDR, negotiate confidentiality and EU terms in an enterprise contract, and do not rely on the default terms for trade secrets or regulated data.
Sources
The same evidence records as the entry sheet. Read means the text was verified; unverified means it is known to exist but not yet read.
Privacy Policyread2026-11-19
Privacy Policy (updated 2025-12-17).
Documentationread2026-11-19
Privacy and security docs: see ev-docs for the default-storage, passthrough and ZDR clauses.
Vendor announcementread2026-11-19
SOC 2 Type II announcement (published 2025-07-08): completed via an independent multi-month audit, auditor not named; HIPAA adherence and business associate agreements stated.
Documentationread2026-11-19
Support knowledge base article, "Last updated 1 year ago": EU data centres exist, and dedicated endpoints on targeted data centres including EU locations are available only on Scale and Enterprise tier plans.
Vendor announcementread2026-11-19
Pricing page: serverless per-token rates with cached-input prices for 24 chat models plus vision, image, video and audio models; on-demand dedicated GPU endpoints (H100 $5.49 per GPU hour); GPU clusters; fine-tuning; provisioned throughput via sales; promotions with end dates.
Documentationread2026-11-19
OpenAI-compatibility docs: an OpenAI-compatible REST API and SDKs across chat, completions, vision, image generation, text-to-speech and speech-to-text, used by pointing an OpenAI client at Together's base URL.
Terms of serviceread2026-11-19
Terms of Service (updated 2026-05-19).
Third-party analysisread2026-11-19
Artificial Analysis provider page: independent third-party analysis of 18 Together AI models by intelligence, speed and price (Kimi K3.1 Pro ranked first), showing the same open-weight models compared across providers.
Data Processing Addendumread2026-11-19
Data Processing Addendum PDF (no effective date on the PDF): Together (US) Inc.; EU SCCs and UK Addendum; the conflict clause prefers the DPA for personal data it covers; 24-hour breach notice (Annex says 48).
Documentationread2026-11-19
Privacy and security and zero-data-retention docs.
Documentationread2026-11-19
Geo-specific inference docs: only a United States endpoint (us-api.together.ai, 1.25x standard rates, 8 models) keeps inference processing in the US.
Vendor announcementread2026-11-19
ISO 27001:2022 blog (published 2026-06-10): Together AI "has received an ISO 27001:2022 certification from A-LIGN Compliance and Security, Inc., an ANAB-accredited certification body", scoped to the ISMS for the global platform including third-party data centres.
Slaread2026-11-19
Public status page (rendered page, index.json and feed): "All services are online".
Securityunverified2026-11-19
Trust centre (trust.together.ai): a Vanta single-page app shell of about 7.9 KB with no readable content.