The Wolf of Defi
DeFi Commentary
Render Network should benefit as exploding GPU demand makes idle GPUs a core part of the solution. Per Render Network, Salad hit a capacity ceiling for the first time in eight years as agentic AI consumes compute at a scale legacy infrastructure wasn't built for, and it is integrating with Render to tap idle GPUs in 180 countries. Salad is just the first of many, and RENDER could see double digits again. x.comÂ
Render Network is the compute layer for AI beyond 3D rendering, because the same GPU that renders a 3D scene runs an AI model, and Render was built to put idle GPUs to work at scale. LTX Studio's chief product officer, Daniel Berkovitz, says AI models are at their core like rendering engines. x.comÂ
GPU DePIN's inference opportunity is far more interesting, but its moat now depends on reliable routing and repeat paying customers rather than incentivized GPU supply. Render Network logged $539K in paid jobs and $512K in RENDER burns over 30 days, though not all jobs were inference and emissions continue. Bittensor subnet rewards do not prove customer demand either. Real utilization, retention, healthy margins and revenue without emissions will show who can sustain the business. x.com x.com … Show more
Customer-funded demand can make decentralized AI compute tokens more valuable; usage alone can hide weak economics. Chutes reported revenue per million tokens up 37.7% and per GPU up 44.9% after cutting free and loss-making traffic, though its gains do not automatically lift TAO. Render's approved Salad roadmap could route payments and rewards through RENDER; integration and recurring customer-funded burns still need proof. Messari reported Akash leases up 27.1% despite revenue falling 45%. … Show more
Render looks overpriced next to Chutes, Bittensor subnet 64: Render carries about a $780M market cap versus Chutes' $77M, making Render the worse buy in 2026. x.comÂ
Render looks like one of the strongest AI infrastructure plays, as accelerating demand for AI compute makes decentralized GPU networks harder to ignore. If AI stays the dominant narrative, the Render Network could have much more room to run, and the chart is starting to wake up. x.comÂ
Lower AI token prices drive more usage, making GPUs scarcer and raising rental rates for neocloud capacity. Gavin Baker reports rising GPU rental prices even as token prices fall; the GPU constraint never left, and higher rental prices make the ARR math work again. x.com x.comÂ
Render turns usage into buy pressure more slowly than Lium Subnet 51 relative to market cap, making Lium the better tokenomics bet. Lium converts GPU rentals into buybacks and emission burns, has already bought back and burned over 1M tokens, and hit new revenue highs. Its owner's tokens are locked in large amounts; the owner is not doxxed, which @DFVTAO says could deter institutional capital, though that is not necessarily bad if intentions are good. x.com x.comÂ
Rising rental rates even for older GPUs strengthen the case that IREN and CoreWeave's compute assets retain value. The quoted a16z data puts year-to-date rental gains at 160% for H200s, 120% for B200s and 35% for H100s; even six-year-old A100s rent for more than at the start of the year. x.comÂ
Rising GPU rental rates and Anthropic's compute margins challenge claims of collapsing demand for CoreWeave's capacity. The WSJ reported Anthropic's first-half revenue of $16.3B against $9B in compute and infrastructure spending; Bloomberg reported its Q3 compute margin above 40%. Long contracts and fully amortizing loans cushion demand declines but cap upside. Despite the challenging neocloud model, CoreWeave's risk/reward is by far the best and its quality well ahead of peers. x.com x.com … Show more
Semiconductor demand could withstand slower OpenAI and Anthropic growth if open-source models and post-training add more chip demand than it removes. Falling token prices need not erode per-GPU economics as cost per token falls and output per GPU rises, potentially keeping rental rates firm. The balance is unresolved; the lean is bullish, but markets may focus on the easier-to-see slowdown. Shares could chop until numbers defeat the bears, or a well-founded bear case could trigger heavy selling. … Show more
Scarce compute prices Nebius's and CoreWeave's GPU rentals by marginal-token value, not chip cost. Better models lift installed watts' value faster than new GPUs devalue old ones. SpaceX says contracts renew at higher dollars per watt; CoreWeave's A100 deal runs to 2029, and Nebius's Blackwell auction cleared 15% above its prior high. The 2-3 year GPU-life bear case keeps getting set back. If supply catches demand, older GPUs would be repriced by performance per watt, but that looks far off. … Show more
