Skip to content
Together AI logo

Together AI

Live·11 models

The widest open-weight catalogue, including models nobody else bothers to host.

Widest open catalogueFine-tuningDedicated endpoints

11 models on Together AI

Together AI's own rate for each of these. Where another provider serves the same model it may charge differently. The model page shows every route side by side.

ModelPriceContext
Multilingual E5 Large Instructintfloat/multilingual-e5-large-instruct$0.020 inFree out1K
FLUX.1 [schnell]black-forest-labs/flux-1-schnell$0.002700 / megapixel
Gemma 3n E4B Instructgoogle/gemma-3n-e4b$0.060 in$0.120 out33K
GPT-OSS 20Bopenai/gpt-oss-20b$0.050 in$0.200 out128K
Qwen3.5 9Bqwen/qwen3.5-9b$0.170 in$0.250 out262K
Llama 3.3 70B Instructmeta/llama-3.3-70b-instruct$1.04 in$1.04 out131K
MiniMax M3minimax/minimax-m3$0.300 in$1.20 out524K
Qwen3.7 Plusqwen/qwen3.7-plus$0.320 in$1.28 out1M
DeepSeek V4 Prodeepseek/deepseek-v4-pro$1.74 in$3.48 out512K
GLM-5.2zai/glm-5.2$1.40 in$4.40 out262K
Kimi K3moonshot/kimi-k3$3.00 in$15.00 out1M

Availability, from our vantage point

What Together AI looked like from Multigrid's servers, not Together AI's own uptime figure, which only they can publish.

99.9% up0.1% degradedReachability probes: 99.91% of 5645, 100% of the window observed.
Probes 99.91%Real traffic 3 requests: too few to publish

The ring is reachability probes, measured across 100% of the last 30 days a share of what we watched, not of the month. We probe when someone loads this page, so the gaps are ours.

  • Succeeded
  • Failed on a degraded day
  • Failed on a bad day (< 50%)
  • Nothing measured yet

The two lanes are never averaged. Probes ask whether Together AI is answering at all; traffic is the share of real requests we routed there that completed. A provider can pass every probe while completing nothing, which is why they are stacked rather than blended, and why a green top lane over a red bottom one is the pattern worth looking for.

Coverage is ours, not theirs: we probe when someone loads the status page, at most once every five minutes, so an empty slot means we were not looking rather than that anything was wrong. For an authoritative account of an incident, read Together AI’s own status page. They are the only ones who can write it.

None of this is latency. What happened to your own requests (timing, errors and cost per provider) is on your analytics page, measured from traffic you actually sent.

Reach Together AI on one balance

We hold a platform key for Together AI, so everything above is reachable on Multigrid credit: one balance across every vendor, at the prices listed. Signing up takes no card and commits you to nothing. If you already have a Together AI contract, bring that key instead and we take no percentage on the traffic.