015Who makes money serving open models?
Serving DeepSeek’s V4.1-Flash on a Blackwell rack earns about $39 an hour per GPU at DeepSeek’s off-peak prices, so a host renting the rack covers its rent at 37% utilization. On an 8-GPU B200 box running NVIDIA’s standard recipe it takes 60%, and an H200 can’t cover its rent at those prices at all. The margin sits with rack-scale serving, frontier-size models and owned GPUs; on the most popular model, 33 hosts compete on price.
What a GPU earns, and the utilization a renter needs
Revenue per GPU-hour at full load. The rack is vLLM’s own figure on SemiAnalysis AgentX traffic. The rest are NVIDIA’s Dynamo recipes on its agent trace (64K-token prompts, 400-token answers, 90% of the prompt cached), priced at each model’s going rate: DeepSeek’s off-peak rate for V4.1-Flash, with the paler bar at the peak rate most OpenRouter hosts post; Z.ai’s list for GLM-5.3-Flash; Moonshot’s $3/$15 for Kimi K3. Ticks are typical on-demand rent on October 20. The right-hand column is the share of hours a renter must sell to cover it.
An owner needs far less. At SemiAnalysis’s owner costs ($1.95 an hour for a B200, $2.65 for a GPU in a GB300 rack) the B200 box covers its costs at 15% utilization and the GB300 rack at 7%. Kimi K3 on H200s doesn’t cover even the cost of owning them.
How the money comes in
Routing is price-first. OpenRouter’s default spreads requests across hosts that have been stable, weighted by the inverse square of price, so a host at half the price draws four times the traffic. It adds no markup to host prices and earns a 5.5% fee when buyers load credits, so a host keeps the price it posts.
The traffic is agents, and mostly input. In the week to October 14, V4.1-Flash processed 36.6 trillion prompt tokens and 807 billion completion tokens, about 45 to 1, and its biggest senders are coding and agent harnesses. In the week to October 26 it carried 52.8T tokens, 28% of everything in OpenRouter’s daily top 50. Chinese models carry 60% of OpenRouter’s tokens over the last four weeks but 26% of the dollars (tracker).
So it’s a cache business priced per input token. On NVIDIA’s trace, 70% of a V4.1-Flash request’s revenue comes from the 10% of prompt tokens that miss the cache; output is 17%. At OpenRouter’s 45-to-1 mix, uncached input is still about 48%. The cache-read price, which hosts set separately, decides what an agent pays. Pack 014 shows the same split from DeepSeek’s side.
Posted isn’t paid. 15 of the 33 hosts list exactly DeepSeek’s peak rate, $0.30 per million input tokens and $1.20 per million output; 8 run discounts of 30% to 84%. OpenRouter now shows next to the posted prices what buyers actually pay on average, which caching and discounts pull well below the list.
What one agent request costs at each host
Dollars per 1,000 requests on NVIDIA’s agent trace (64K-token prompt, 90% of it cached, 400-token answer) at each host’s cheapest standard endpoint for V4.1-Flash on October 27, discounts included. Listed companies in orange. Dashed lines: DeepSeek’s own off-peak and peak rates. Two hosts are left out: one lists no cache price and one listing couldn’t be read.
Headline prices mislead. Decart has the 2nd-lowest input-plus-output price but ranks 11th on an agent’s bill, because its cache reads cost $0.018 per million, three times DeepSeek’s peak rate. Alibaba Cloud drops from 10th to 19th the same way. Modal and Google Vertex charge $0.03 for cache reads and cost 3.0× DeepSeek’s off-peak rate per request.
Which models pay best
Frontier-size open models, on the newest racks. Kimi K3 has 2.8 trillion parameters, 104 billion of them active, and nine of the ten platforms that served it in its first weeks charged Moonshot’s $3 input, $15 output. That price holds because the economics only work at rack scale: on NVIDIA’s recipes a GB300 GPU earns $34.91 an hour on K3, a GB200 $12.58 and an H200 $1.35.
New models, for a few days. StepFun’s Step 5 Preview reached second place on OpenRouter in its first week, and K3 was live on nearly every major platform within 48 hours of its weights. The window before thirty hosts pile in is short.
The Flash tier is where the volume is. V4.1-Flash, GLM-5.3-Flash and MiniMax’s M3.1-Flash carry the most tokens, and their prices follow cost down: over about 90 days the cheapest input price fell 77% for V4.1-Flash and 90% for GLM-5.3-Flash. GLM-5.3-Flash at Z.ai’s list earns $15.05 an hour on a B200, slightly more than V4.1-Flash at DeepSeek’s off-peak rate. Margins in this tier go to whoever runs the rack-scale stack.
Hopper loses on every model here. An H200 can’t cover its rent on V4.1-Flash at DeepSeek’s off-peak prices, and NVIDIA doesn’t publish an H100 recipe for the model at all. Free endpoints, MiniMax’s M3.1-Flash and NVIDIA’s Nemotron 3 Ultra among them, carried 12.6T and 5.5T tokens in the week to October 26, free to users, so someone else pays.
Who keeps the margin
Reported figures. The measures differ, so compare rows with care.
| Company | Layer | Measure | Value | Note |
|---|---|---|---|---|
| DeepSeek | Lab serving its own open weights | Theoretical cost-profit ratio, V3 and R1 on H800s priced at $2 an hour · Feb 2025 | 545% | At R1 list prices; DeepSeek said actual revenue was far lower |
| NebiusNBIS | GPU cloud with a token API | Adjusted EBITDA margin, AI business · Q2 2026 | 50% | Mostly GPU rental; Token Factory is its fastest-growing line |
| Fireworks | Inference platform, private | Gross margin · 2026 | ~50% | Targeting 60% |
| CerebrasCBRS | Inference cloud on its own chips | Core gross margin · Q2 2026 | 41% | Cloud services 53% (core) in Q1 |
| Zhipu2513.HK | Lab selling its open weights by API | Gross margin on cloud API · H1 2026 | 24.6% | −0.4% a year earlier, after raising API prices about 101% this year |
| MiniMax0100.HK | Lab selling its open weights by API | Gross margin, whole company · H1 2026 | 17.9% | Open-platform revenue up 703% |
How the margin moved
2023 to 2025: compute got cheap. H100 rents fell from peaks near $8 an hour to $2.00 on Silicon Data’s index by December 2025. In February 2025 DeepSeek said its V3 and R1 fleet cost $87,072 a day at $2 an H800-hour against $562,027 of theoretical revenue, about $12.90 per GPU-hour, while warning that actual revenue was far lower.
2026: compute got scarce again. H100 and H200 contract prices rose about 40% between October 2025 and March 2026 as memory costs passed through. Silicon Data’s H100 index reached $2.88 on October 24, and Ornn’s settled at $3.17 on October 26, up 24.8% in a month.
Token prices split. The Flash tier fell on architecture: V4.1-Flash activates 8B parameters to read and 16B to write. The frontier tier went up: K3 launched at $3/$15, Zhipu raised API prices about 101% this year, and DeepSeek warned of a significant price increase. Revenue per GPU-hour on Blackwell racks is about three times DeepSeek’s 2025 figure even after its price cuts (pack 014), so hosts running the rack-scale stack kept ahead of rising rents and hosts on smaller setups fell behind.
The clearest public read is Zhipu: its API sales went from a −0.4% gross margin to 24.6% in a year, with prices about doubled. Selling tokens is still a thin business for anyone without the cheapest capacity. Owning the rack, or the model, is where the margin is.
If the GPU curve inverts
GPU rents aren’t inverted now. Silicon Data says its modeled forward curve sloped down in 2025 and moved to contango this year. In mid-2026 its B200 curve was only about 8% lower at 36 months ($5.62 to $5.17), and H100 was flat past 15 months. CME has distributed that curve to subscribers since October 15, and the H100 and B200 futures it planned to list on October 5 were reported as still pending regulatory review. Our own index (tracker) started this week and doesn’t yet have enough term quotes to read a curve.
A steep inversion, a high near month and a low back end, would mean scarce capacity now and a supply wave or fast obsolescence later. A GPU-hour can’t be stored, so, as with power or freight, nothing ties the two ends together. Who gains, and who loses:
Winners and losers in a steeply inverted curve
Our assessment, roughly in order of how much each gains.
| Owners of installed GPUs that aren’t under contract | They sell at the expensive near end and stay off the cheap back end. Their cash margin is the spot rent less power. |
| Hosts that own their GPUs | Token prices settle where hosts renting at prompt rates break even. Owners keep the difference, and the inversion widens it. |
| Buyers who can wait | Training runs, RL and batch jobs buy the cheap back end. A long position in a deferred future gains as it rolls up toward spot, if spot holds. |
| Holders of leases fixed near the 2025 low | They are long compute below market and can undercut renters or sublet where contracts allow. |
| Capacity that can switch on now | Energized sites and the bottleneck parts, such as HBM. An inverted curve pays for speed. |
| Brokers and market makers | Steep curves and wide gaps between chips, regions and indices bring hedging demand and spreads to capture. |
| New builds that need contracts to borrow | Lower contracted rates mean less debt per GPU, which slows new supply and pulls the curve back. |
| Hosts renting at the prompt price | They pay the scarce price for GPUs and sell tokens at prices set by owners. |
| Long-dated bets on resale value | An inverted curve is the market pricing faster depreciation than six-year accounting lives. |
How new hosts get in
Most of the 33 hosts on V4.1-Flash fit four groups: inference-engine and model companies (Modular, Relace, Morph, Decart, Reka, Cohere), GPU clouds selling tokens as well as GPU-hours (CoreWeave, Nebius, DigitalOcean, Crusoe, GMI, Prime Intellect), Chinese clouds (Baidu, Alibaba, SiliconFlow), and privacy-focused hosts (Phala, Venice).
Uptime first. Over the day to October 27, OpenRouter served V4.1-Flash 99.91% of the time; without its failover between hosts it would have been 87.81%. Hosts that stay up get the traffic. Then price: with inverse-square routing, a launch discount buys share fast. Pass the quality checks: OpenRouter publishes each host’s benchmark scores and has a routing mode for tool-calling accuracy, so heavy quantization shows. Be early, or fast: day-zero support catches a model before prices converge, and a speed tier sells at a premium (Baseten Fast posts $0.60/$2.40).
The arithmetic sets the floor. A host renting 8-GPU B200 boxes and pricing at DeepSeek’s off-peak rate needs 60% utilization to cover its rent. The hosts that last own their GPUs, fill idle ones, run the rack-scale stack, or have someone else paying.
Listed ways in
No listed company only rents GPUs and resells open-model tokens. The listed exposure is GPU clouds with token APIs on their own GPUs, Cerebras on its own chips, and the Hong Kong-listed labs selling their own open weights.
| Company | What it sells | Latest figures |
|---|---|---|
| NebiusNBIS | Token Factory: per-token API on 60+ open models, on its own GPUs. Hosts V4.1-Flash on OpenRouter at $0.30/$1.20. | Q2 revenue $582M; AI business 50% adjusted EBITDA margin; bought Eigen AI for $643M |
| CoreWeaveCRWV | Serverless Inference: per-token API on an open-model catalog, on its own GPUs. Hosts V4.1-Flash on OpenRouter at $0.15/$0.65. | Managed-inference booked ARR went from $1M to over $100M within months of launch |
| CerebrasCBRS | Fast inference cloud on its own wafer-scale chips. | IPO in May raised $6.4B; Q2 core cloud revenue $127.7M, up 287%; core gross margin 41% |
| DigitalOceanDOCN | Hosts V4.1-Flash ($0.225/$0.90) and Kimi K3 on OpenRouter. | — |
| Zhipu2513.HK | Sells its own open-weight GLM models by API, including through OpenRouter. | API revenue RMB 825M in H1, up 2,736%, 86.5% of revenue; API gross margin 24.6% |
| MiniMax0100.HK | Sells its own M-series models by API, including through OpenRouter. | Open-platform revenue up 703% in H1; gross margin 17.9% |
| Alphabet, Alibaba, BaiduGOOGL · BABA · BIDU | Host V4.1-Flash on OpenRouter: Vertex at the posted rate, Alibaba Cloud at 50% off, Baidu Qianfan at 84% off. | — |
The pure-play hosts are private: Fireworks (over $1B annualized revenue, valued at $17.5B in July), Together ($8.3B), Baseten ($13B) and DeepInfra. Lambda, a GPU cloud, is raising up to $4B before a possible IPO at a $14.5B pre-money valuation.
What this is anchored on
Host prices as posted on October 27; rents as of October 20 (pack 014) and the dates shown.
| Item | Data point | Source |
|---|---|---|
| Rack-scale yield | $38.96 per GPU-hour at DeepSeek's off-peak prices on one GB300 NVL72 rack, SemiAnalysis AgentX traffic; 954K tokens/s per GPU counting cached input | vLLM blog, Oct 23 |
| Recipe yields | Output tokens/s per GPU on the agentic trace: V4.1-Flash 1,155 (GB200), 1,088 (B200), 359 (H200), disaggregated | NVIDIA Dynamo, DeepSeek-V4.1-Flash recipe |
| Recipe yields | Kimi K3: 91.3 (GB300), 32.9 (GB200), 3.54 (H200) output tokens/s per GPU; trace: 64K input, 400 output, 90% cache hits | NVIDIA Dynamo, Kimi-K3 recipe |
| Recipe yields | GLM-5.3-Flash: 579 output tokens/s per GPU on four B200s | NVIDIA Dynamo, GLM-5.3-Flash recipe |
| Hosts | 33 hosts for V4.1-Flash with input, output and cache-read prices; pooled availability 99.91% against 87.81% without routing; top apps are agent harnesses | OpenRouter, DeepSeek V4.1-Flash |
| Routing | Default routing weights stable hosts by the inverse square of price | OpenRouter docs, provider routing |
| Fees | No markup on host prices; 5.5% fee on credit purchases | OpenRouter FAQ |
| Traffic mix | V4.1-Flash, week to Oct 14: 36.62T prompt and 807B completion tokens; Step 5 Preview reached #2 in its first week | Metaprompt, from OpenRouter's weekly table |
| Traffic mix | Open-weight models carry about 60% of tokens from US-originating requests on OpenRouter, mostly Chinese models | The New Stack |
| Prices | V4.1-Flash off-peak $0.15 input, $0.003 cached, $0.60 output; peak hours double | Apidog |
| Prices | Kimi K3: 2.8T parameters, 104B active; providers' prices | DeepInfra |
| Prices | Nine of ten K3 platforms at $3/$15 with $0.30 cached; most live within 48 hours of the weights | DEV Community |
| Prices | GLM-5.3-Flash $0.15 input, $0.03 cached, $0.50 output | OrcaRouter |
| Prices | DeepSeek warned of a significant price increase after launching V4 at very low rates | VKTR |
| Price falls | V4.1-Flash cheapest input price down 76.7% over about 90 days, $0.30 to $0.07 | PricePerToken |
| Price falls | GLM-5.3-Flash cheapest input price down 90% over about 90 days | PricePerToken |
| Rents | H100 rents have come down from historical peaks near $8 an hour | Thunder Compute |
| Rents | H100 $3.17 an hour on Oct 26, up 24.8% in a month | Ornn Compute Price Index |
| Rents | Silicon Data H100 index $2.00 (Dec 9, 2025) to $2.88 (Oct 24, 2026); its forward curve moved from sloping down to contango | FourWeekMBA |
| Rents | Mid-2026 forward curve: B200 about 8% lower at 36 months; H100 flat past 15 months | Silicon Data |
| Rents | H100 and H200 contract prices up about 40% from Oct 2025 to Mar 2026 as HBM costs passed through | IntuitionLabs |
| Futures | CME distributes Silicon Data's forward curves from Oct 15; H100 and B200 futures planned for Oct 5, pending regulatory review | Crypto Briefing; Spheron |
| Owner cost | Owner cost per GPU-hour: B200 $1.95, GB200 $2.21, GB300 $2.65 | SemiAnalysis via InferenceX |
| Margins | DeepSeek's Feb 2025 disclosure: $87,072 a day of GPU cost against $562,027 of theoretical revenue | South China Morning Post |
| Margins | Zhipu cloud API gross margin −0.4% to 24.6%; API prices up about 101% this year | 36Kr |
| Margins | Zhipu API revenue and share; Zhipu and MiniMax gross margins, H1 2026 | 36Kr |
| Margins | MiniMax open-platform revenue up 703.1% | MiniMax interim results, HKEX |
| Margins | Fireworks gross margin about 50%, targeting 60% | Sacra |
| Companies | Cerebras Q2: core cloud revenue $127.7M, core gross margin 41%, $6.4B IPO | Cerebras 8-K |
| Companies | Nebius Q2 revenue $582M; AI business 50% adjusted EBITDA margin | Zacks via CNHI |
| Companies | CoreWeave managed-inference booked ARR from $1M to over $100M | Futurum |
| Companies | Fireworks, Together and Baseten valuations and revenue | VC Cafe |
| Companies | Lambda raising up to $4B at $14.5B pre-money | Let's Data Science |
Method
Revenue per GPU-hour. For NVIDIA’s recipes: output tokens per second per GPU ÷ 400 output tokens per request × dollars per request × 3,600. Dollars per request: 64,000 prompt tokens, 10% at the uncached input price and 90% at the cache-read price, plus 400 output tokens. Full load means the recipe’s measured concurrency around the clock. NVIDIA’s output figures include reasoning tokens; its GLM and Kimi runs use a 15% subset of the trace and a simulated speculative-decoding acceptance rate.
The rack is vLLM’s figure on one GB300 NVL72 at DeepSeek’s off-peak prices, on SemiAnalysis’s AgentX traffic, a different trace from NVIDIA’s. The paler bars double the V4.1-Flash figures, since DeepSeek’s peak rates double every price.
Break-even utilization is rent ÷ revenue at full load. Rents are typical on-demand prices on October 20, as in pack 014; committed rents are lower, so these overstate what a long-term renter needs. Owner costs are SemiAnalysis’s total cost of ownership from its 2025 survey.
Not counted: CPUs, networking, cache storage, staff, a GPU cloud’s own margin on top of the GPU, and capacity sitting idle between peaks. NVIDIA’s trace is more input-heavy than OpenRouter’s 45-to-1 mix.
Hosts are each host’s cheapest standard-routing endpoint on OpenRouter’s V4.1-Flash page on October 27, at discounted prices where a discount runs; list prices are kept in the data. Token counts for the week to October 26 are from our OpenRouter tracker’s daily top 50.