# Together AI Data Centers, Capex and GPU Pricing Source: https://savrn.com/ai-index/companies/together-ai Markdown alternate of the page above; the site index is https://savrn.com/llms.txt --- [SAVRN Index.](https://savrn.com/ai-index) Explore the IndexCompanies Together AI raised a $305 million Series B in February 2025 led by General Catalyst with Prosperity7, reporting 200MW of secured capacity and a 36,000 GB200 NVL72 deployment with Hypertec, followed by an $800 million Series C in July 2026 with Aramco Ventures, Nvidia and Vista at an $8.3 billion valuation, citing more than 500MW of capacity commitments. Named anchor customer contract values are n.a. Updated October 8, 2026Capital record from the Capital Atlas Named sites 0 Site commitments on record for Together AI Disclosed site capital n.a. Named sites with a stated dollar figure. A floor, not a capex total Disclosed capacity n.a. Named sites with a stated megawatt figure Financing transactions 3 Transactions naming Together AI or one of its sites Prices tracked 12 9 confirmed on the provider's feed or page at the last check Sources 6 Every cited page, listed below Market Together AI is privately held, so there is no public market data to show. Spending ## Reported and guided spending No reported or guided spending line is on record for Together AI. Its position shows in the sites, transactions and facts below. Named sites ## Where Together AI builds No named site commitment is on record for Together AI as of October 8, 2026. Its position shows in the transactions and facts below. Financing ## Financing transactions Transactions that name Together AI as a party or attach to one of its named sites. Amounts are as stated, and overlapping commitments are never summed. | Parties | Instrument | Amount | Date | Site | Sources | | --- | --- | --- | --- | --- | --- | | Together AI + Aramco Ventures + Nvidia + VistaSeries C citing more than 500MW of capacity commitments. | equity raise | $800M | July 1, 2026 | n.a. | | | Together AI + General Catalyst + Prosperity7Series B led by General Catalyst with Prosperity7, reporting 200MW of secured capacity and a 36,000 GB200 NVL72 deployment with Hypertec. | equity raise | $305M | February 20, 2025 | n.a. | | | IBM + Together AI + NvidiaIBM agreed to deploy a large cluster of Nvidia HGX B300 systems for Together AI under a $240 million infrastructure deal. The hardware is expected to become available through IBM Cloud in the first half of 2027. Together AI will use the cluster to host workloads for customers of its public cloud | infrastructure capacity agreement | $240M | Announced 11 August 2026 | n.a. | | The record ## Sourced facts on Together AI - Together AI reported 200MW of secured capacity and a 36,000 GB200 NVL72 deployment with Hypertec at the Series B, rising to more than 500MW of capacity commitments at the Series C. - Named anchor customer contract values are n.a. in the fetched record. Against the market ## How Together AI prices compare Together AI's lowest price for each GPU beside the lowest price every other tracked provider lists for the same GPU, sold the same way. Typical is the middle provider, one price each. ### NVIDIA H100On demand [All NVIDIA H100 prices →](https://savrn.com/ai-index/pricing/gpus/h100) Together AI$3.99per GPU-hour, on-demand Rank10 of 216% below typical Typical$4.25middle of 21 providers Lowest anywhere$1.99Voltage Park Voltage Park1. H100 (single GPU, no contract) $1.99 GMI Cloud2. NVIDIA H100 $2.00 Vultr3. NVIDIA HGX H100 — Cloud GPU $2.30 Denvr Dataworks4. NVIDIA H100 SXM (8 GPUs, 208 vCPU) $2.45 Cirrascale5. 8X NVIDIA H100 (16-Node+ Clustered) $2.49 Massed Compute6. H100 (80 GB) ×1/×2/×4 $2.73 Runpod7. H100 PCIe / H100 NVL — Pods $2.89 Hyperbolic8. H100 SXM $3.19 Lambda9. NVIDIA H100 PCIe $3.29 Together AI10. H100 (80 GB), 8–256 GPUs $3.99 Together AITypical $4.2511 more providers on the [NVIDIA H100 page](https://savrn.com/ai-index/pricing/gpus/h100) ### NVIDIA H100Reserved [All NVIDIA H100 prices →](https://savrn.com/ai-index/pricing/gpus/h100) Together AI$3.19per GPU-hour, reserved Rank3 of 97% below typical Typical$3.42middle of 9 providers Lowest anywhere$2.30Vultr Vultr1. NVIDIA HGX H100 — Bare Metal (8 × H100 80 GB SXM) $2.30 Impossible Cloud2. H100 8-GPU nodes (EU, offer 301-H100) $2.63 Together AI3. reserved $3.19 DigitalOcean4. 12-month reserved $3.26 Cirrascale5. 8X NVIDIA H100 (Standalone) $3.42 Astra Intellectus6. H100 SXM 8-GPU $3.65 Microsoft Azure7. Standard_ND96is_H100_v5 (no InfiniBand variants incl. _noIB_, _flex_) $4.86 Google Cloud8. 3-year committed use $4.86 Lambda9. 1-Click Clusters — NVIDIA H100 $5.54 Together AITypical $3.42 ### NVIDIA H200On demand [All NVIDIA H200 prices →](https://savrn.com/ai-index/pricing/gpus/h200) Together AI$5.99per GPU-hour, on-demand Rank10 of 168% above typical Typical$5.56middle of 16 providers Lowest anywhere$2.60GMI Cloud GMI Cloud1. NVIDIA H200 $2.60 Massed Compute2. H200 NVL (141 GB) ×8 $3.62 Hyperbolic3. H200 $3.99 Crusoe Cloud4. NVIDIA H200 $4.29 Runpod5. Clusters — H200 SXM $4.31 DigitalOcean6. NVIDIA HGX H200 (141 GB) $4.47 Cirrascale7. 8X H200 $4.54 Nebius8. NVIDIA HGX H200 $5.40 Verda9. 1x H200 SXM5 141GB $5.72 Together AI10. H200 (140 GB), 256–1,000 GPUs $5.99 Together AITypical $5.566 more providers on the [NVIDIA H200 page](https://savrn.com/ai-index/pricing/gpus/h200) ### NVIDIA H200Reserved [All NVIDIA H200 prices →](https://savrn.com/ai-index/pricing/gpus/h200) Together AI$3.99per GPU-hour, reserved Rank6 of 810% above typical Typical$3.64middle of 8 providers Lowest anywhere$3.05Impossible Cloud Impossible Cloud1. H200 SXM 8-GPU nodes (EU, offer 313-H200) $3.05 TheAI Cloud2. 1-year reserved $3.20 DigitalOcean3. 12-month reserved $3.40 Cirrascale4. 8X NVIDIA H200 $3.63 Astra Intellectus5. H200 SXM 8-GPU $3.65 Together AI6. reserved $3.99 Google Cloud7. 3-year committed use $4.65 Microsoft Azure8. Standard_ND96isr_H200_v5 (ND H200 v5) $5.28 Together AITypical $3.64 ### NVIDIA B200On demand [All NVIDIA B200 prices →](https://savrn.com/ai-index/pricing/gpus/b200) Together AI$8.19per GPU-hour, on-demand Rank10 of 161% above typical Typical$8.08middle of 16 providers Lowest anywhere$3.50Vultr Vultr1. NVIDIA HGX B200 — Cloud GPU $3.50 GMI Cloud2. NVIDIA B200 $4.00 Massed Compute3. B200 SXM6 ×8 $5.43 Hyperbolic4. B200 $5.99 Cirrascale5. 8X B200 $5.99 Lambda6. NVIDIA B200 SXM6 $6.69 Runpod7. B200 (180 GB, 28 vCPU) — Pods $7.99 Google Cloud8. a4-highgpu-8g (A4) $8.06 Verda9. Instant clusters, B200 SXM6 180GB $8.11 Together AI10. HGX B200 (180 GB) $8.19 Together AITypical $8.086 more providers on the [NVIDIA B200 page](https://savrn.com/ai-index/pricing/gpus/b200) ### NVIDIA B200Reserved [All NVIDIA B200 prices →](https://savrn.com/ai-index/pricing/gpus/b200) Together AI$6.79per GPU-hour, reserved Rank4 of 617% above typical Typical$5.79middle of 6 providers Lowest anywhere$3.50Vultr Vultr1. NVIDIA HGX B200 — Bare Metal (8 × B200 192 GB) $3.50 TheAI Cloud2. 24-month reserved $4.50 Cirrascale3. 8X NVIDIA B200 $4.79 Together AI4. reserved $6.79 Google Cloud5. 3-year committed use $7.09 Lambda6. 1-Click Clusters — NVIDIA HGX B200 $8.87 Together AITypical $5.79 ### NVIDIA B300On demand [All NVIDIA B300 prices →](https://savrn.com/ai-index/pricing/gpus/b300) Together AI$9.99per GPU-hour, not published Rank5 of 9In line with typical Typical$9.99middle of 9 providers Lowest anywhere$6.60Massed Compute Massed Compute1. B300 SXM6 ×8 (280 vCPU, 2800 GB) $6.60 TheAI Cloud2. B300 (8× B300 SXM6 · 288GB … whole node) $6.90 Nebius3. NVIDIA HGX B300 $7.85 Runpod4. B300 (288 GB HBM3e, 32 vCPU) — Pods $8.99 Together AI5. HGX B300 (270 GB) $9.99 Verda6. 1x B300 SXM6 268GB $10.38 AWS7. p6-b300.48xlarge $14.04 Fireworks AI8. B300 288 GB $15.00 Oracle OCI9. BM.GPU.B300.8 $15.00 Together AITypical $9.99 Tracked prices ## Together AI prices Every price the Index tracks for Together AI, in the provider's own words beside a normalized figure, with the date it was last checked. [What AI Costs](https://savrn.com/ai-index/pricing) compares them across the market. | GPU | Offer | Terms | Published price | Per GPU-hour | Change | Checked | Source | | --- | --- | --- | --- | --- | --- | --- | --- | | [NVIDIA H100](https://savrn.com/ai-index/pricing/gpus/h100) | H100 (80 GB), 8–256 GPUs | on-demandList | $3.99/hr per GPU, on-demand | $3.99 | No change | On pageOct 24, 2026 | [together.ai ↗](https://www.together.ai/gpu-clusters) | | [NVIDIA H100](https://savrn.com/ai-index/pricing/gpus/h100) | reserved | reservedCommitted | Starting at $3.19/hr per GPU, reserved | $3.19 | No change | On pageOct 24, 2026 | [together.ai ↗](https://www.together.ai/gpu-clusters) | | [NVIDIA H200](https://savrn.com/ai-index/pricing/gpus/h200) | H200 (140 GB), 256–1,000 GPUs | on-demandList | $5.99/hr per GPU, on-demand | $5.99 | No change | On pageOct 24, 2026 | [together.ai ↗](https://www.together.ai/gpu-clusters) | | [NVIDIA H200](https://savrn.com/ai-index/pricing/gpus/h200) | reserved | reservedCommitted | Starting at $3.99/hr per GPU, reserved | $3.99 | No change | On pageOct 24, 2026 | [together.ai ↗](https://www.together.ai/gpu-clusters) | | [NVIDIA B200](https://savrn.com/ai-index/pricing/gpus/b200) | HGX B200 (180 GB) | on-demandList | $8.19/hr per GPU, on-demand | $8.19 | No change | On pageOct 24, 2026 | [together.ai ↗](https://www.together.ai/gpu-clusters) | | [NVIDIA B200](https://savrn.com/ai-index/pricing/gpus/b200) | reserved | reservedCommitted | Starting at $6.79/hr per GPU, reserved | $6.79 | No change | On pageOct 24, 2026 | [together.ai ↗](https://www.together.ai/gpu-clusters) | | [NVIDIA GB200 NVL72](https://savrn.com/ai-index/pricing/gpus/gb200) | GB200 NVL72 (186 GB), 512–1,000+ GPUs | not publishedNot published | , | Not published | No change | On pageOct 24, 2026 | [together.ai ↗](https://www.together.ai/gpu-clusters) | | [NVIDIA B300](https://savrn.com/ai-index/pricing/gpus/b300) | HGX B300 (270 GB) | List | $9.99/hr per GPU, on-demand | $9.99 | No change | On pageOct 24, 2026 | [together.ai ↗](https://www.together.ai/gpu-clusters) | | [NVIDIA GB300 NVL72](https://savrn.com/ai-index/pricing/gpus/gb300) | GB300 NVL72 (288 GB) | not publishedNot published | , | Not published | No change | On pageOct 24, 2026 | [together.ai ↗](https://www.together.ai/gpu-clusters) | | Seller | Model | Input | Cached input | Output | [Revenue per GPU-hour at list](https://savrn.com/ai-index/pricing/gpus/vera-rubin#token-value) | Other published prices | Change | Checked | | --- | --- | --- | --- | --- | --- | --- | --- | --- | | Together | Kimi K3 | $3.00 | n.a. | $15.00 | n.a. | None published | No change | ResearchOct 21, 2026[together.ai ↗](https://www.together.ai/pricing) | | Together | DeepSeek V4 Pro 0813 | $1.32 | n.a. | $3.96 | n.a. | None published | No change | ResearchOct 21, 2026[together.ai ↗](https://www.together.ai/pricing) | | Together | gpt-oss-120B | $0.15 | n.a. | $0.60 | n.a. | None published | No change | ResearchOct 21, 2026[together.ai ↗](https://www.together.ai/pricing) | Revenue per GPU-hour is SAVRN's calculation, not a price any seller charges: the 6.83 million input and 0.85 million output tokens one Vera Rubin GPU handles in an hour in CoreWeave's measured test, billed at this model's list prices. It puts a token price beside a GPU rental price; it says nothing about how fast this model actually runs. Sources ## Sources for Together AI Every source cited on this page, once each. - together.ai[Together AI](https://www.together.ai/blog/announcing-our-series-c) - techcrunch.com[TechCrunch](https://techcrunch.com/2026/07/01/neocloud-together-ai-raises-800m-leaps-to-8-3b-valuation/) - together.ai[Together AI](https://www.together.ai/blog/together-ai-announcing-305m-series-b) - siliconangle.com[SiliconANGLE](https://siliconangle.com/2026/08/11/ibm-inks-240m-infrastructure-deal-ai-optimized-cloud-operator-together-ai/) - together.ai[together.ai](https://www.together.ai/gpu-clusters) - together.ai[together.ai](https://www.together.ai/pricing) NeoCloud ## More in the Index [Applied Digital](https://savrn.com/ai-index/companies/applied-digital)[Bitdeer Technologies Group](https://savrn.com/ai-index/companies/bitdeer)[Cerebras Systems](https://savrn.com/ai-index/companies/cerebras)[Cipher Digital Inc.](https://savrn.com/ai-index/companies/cipher-digital)[Core Scientific](https://savrn.com/ai-index/companies/core-scientific)[CoreWeave](https://savrn.com/ai-index/companies/coreweave)[Corvex, Inc.](https://savrn.com/ai-index/companies/corvex)[Crusoe Energy](https://savrn.com/ai-index/companies/crusoe-energy)[Fluidstack](https://savrn.com/ai-index/companies/fluidstack)[Galaxy Digital Inc.](https://savrn.com/ai-index/companies/galaxy-digital)[Groq](https://savrn.com/ai-index/companies/groq)[Hut 8 Corp.](https://savrn.com/ai-index/companies/hut8)[IREN](https://savrn.com/ai-index/companies/iren)[Lambda (Lambda Labs)](https://savrn.com/ai-index/companies/lambda)[Nebius Group](https://savrn.com/ai-index/companies/nebius-group)[Nscale](https://savrn.com/ai-index/companies/nscale)[Poolside](https://savrn.com/ai-index/companies/poolside)[SambaNova Systems](https://savrn.com/ai-index/companies/sambanova)[Sesterce](https://savrn.com/ai-index/companies/sesterce)[TeraWulf](https://savrn.com/ai-index/companies/terawulf)[Voltage Park](https://savrn.com/ai-index/companies/voltage-park)[Vultr](https://savrn.com/ai-index/companies/vultr) Capital figures come from the [Capital Atlas](https://savrn.com/data-center-capital-atlas), refreshed every Tuesday, Thursday and Saturday. Site commitments overlap reported capital expenditure and overlap one another, so each figure appears as its source states it and none are added into a single deployment number. Informational, not investment advice. Sources The SAVRN reference set ## Seven references, one system Capital, delay, policy, water, grid, permits and scarcity are one system seen from seven sides. Each reference is primary-sourced, refreshed on a schedule, and links into the others, so a project can be followed from the money committed to the docket that stopped it. Capital Atlas Who is funding AI data centers: capital providers, named sites, financing transactions, and the blocked-and-delayed record, state by state and company by company. [Open the atlas →](https://savrn.com/data-center-capital-atlas) Delay Watchlist Project-level slippage: named projects, jurisdictional actions, utility and federal dockets, SEC disclosures, and the interconnection queue. Open the watchlist → Moratorium Tracker The policy layer: every state measure and local moratorium mapped and cited, with one page per state. Open the tracker → Water Tracker Every federal, state, local and interstate water measure that touches a data center, plus a fifty-state water ledger. Open the water tracker → Grid Operator Watchlist ERCOT, PJM, MISO, CAISO, SPP, NYISO, ISO-NE and FERC: the large-load rules, filings and queue positions that decide whether a site can energize. Open the watchlist → Permits and Power Development Data center and substation permits, large loads in the interconnection queues, and the power plants planned to serve them, refreshed hourly from public records. Open the permits tracker → SAVRN Scarcity Tracker Eleven primary-sourced indicators of AI infrastructure scarcity, one composite index, read against the token index, SAVRN's own cost basis and the frontier ceiling. Open the scarcity tracker → Have a specific address? Site Intelligence reads all seven references against one location, county, utility and grid operator included, and cites every finding. [Check an address →](https://savrn.com/site-intelligence)