Subnet teams' own channels, X posts, press, newsletters and events read from the chain — GitHub lives in its own tab
Cacheon@cacheon_aiWhy DeepSeek-V4.1-Flash for our next competition? Because targeting a frontier open model keeps Cacheon at the leading edge. Strong coding and agent performance. Native visual understanding. An architecture designed for efficiency. How much faster can we run DeepSeek-V4.1-Flash on the same hardware?
Validator hotkey 5CoZxgtfhcJKX2HmkwnsN18KbaT9aih9eF3b6qVPTgAUbifj announced new child keys on SN1, SN2, SN3, SN4, SN5, SN6, SN7, SN8, SN9, SN10, SN11, SN12 and 113 more, applying at block 9,281,605 (~24h). Stake delegated to it there routes through its children after that.
Between blocks 9,274,365 and 9,274,433
Cacheon@cacheon_aiA little summary of our performance progress so far on #SN14. Speeds are measured using AIPerf which simulates multi-user concurrent workloads. Can you do better? Submit your improved kernel and earn SN14-Alpha on Bittensor $TAO Need a speed-boost? Reach out and we'll tailor to your needs.
Validator hotkey 5E4z3h9yVhmQyCFWNbY9BPpwhx4xFiPwq3eeqmBgVF6KULde announced new child keys on SN1, SN2, SN3, SN6, SN7, SN8, SN9, SN11, SN12, SN13, SN14, SN17 and 6 more, applying at block 9,265,930 (~24h). Stake delegated to it there routes through its children after that. It replaces the changes that were pending on SN1 (block 9,265,620), SN2 (block 9,265,587), SN3 (block 9,265,590), SN6 (block 9,265,419), SN7 (block 9,265,422), SN8 (block 9,265,425) and 12 more, which will not apply then.
Between blocks 9,258,715 and 9,258,782
Validator hotkey 5E4z3h9yVhmQyCFWNbY9BPpwhx4xFiPwq3eeqmBgVF6KULde announced new child keys on SN1, SN5, SN6, SN7, SN8, SN9, SN10, SN11, SN12, SN13, SN14, SN84 and 9 more, applying at block 9,265,380 (~24h). Stake delegated to it there routes through its children after that.
Between blocks 9,258,178 and 9,258,244
Validator hotkey 5CqRMuYSSqYkSa8AM3VawVfdyUhyBufyesvvgh3vCgaGb9Ed announced new child keys on SN14, SN78, SN93, applying at block 9,262,861 (~24h). Stake delegated to it there routes through its children after that.
Between blocks 9,255,659 and 9,255,785
Cacheon@cacheon_aiAffine's King model. Two weeks of optimization. 40.9% more throughput. @affine_io became Cacheon’s first customer. These are the performance numbers behind that partnership. We pushed Qwen3.6-35B to 528.9 tokens/s, up from 375.4 tokens/s on stock SGLang. We’re building on this work as we bring Cacheon’s inference optimization to more customers and models.
Cameron Fairchild@KibibyteMeVery happy with where #SN14 is heading. Excited to work with more teams in the eco. Lots of synergies and probably the best pool of customers.
Quoting @cacheon_ai
$100k in revenue last month. @affine_io became our first customer. We’re now working to monetize our Qwen and GLM-5.3 optimizations through licensing deals and our own inference service. This quarter, we’re targeting more customers and expanding competitions to high-volume models - building a catalogue of kernels we can license or deploy ourselves. There's more to come. Keep calm and Cacheon.
Cacheon@cacheon_ai$100k in revenue last month. @affine_io became our first customer. We’re now working to monetize our Qwen and GLM-5.3 optimizations through licensing deals and our own inference service. This quarter, we’re targeting more customers and expanding competitions to high-volume models - building a catalogue of kernels we can license or deploy ourselves. There's more to come. Keep calm and Cacheon.
Cacheon@cacheon_aiQ3 is over. How are we doing against our roadmap? The proof of concept is built: ✓ Multiple models optimized - MiniMax-M3., GLM-5.3 and Qwen3.6. ✓ Competitions run in parallel across models. ✓ Speed improvements evaluated end to end, with output-quality checks. Different models suit different jobs. This progress opens a path to serving more workloads, with improvements judged on both speed and quality. Q4’s focus: monetization and product–market fit.
Validator hotkey 5GP7c3fFazW9GXK8Up3qgu2DJBk8inu4aK9TZy3RuoSWVCMi announced new child keys on SN14, applying at block 9,235,932 (~24h). Stake delegated to it there routes through its children after that.
Between blocks 9,228,708 and 9,228,769View subnet →
Cacheon@cacheon_aiAI inference is getting cheaper. But it’s consuming more and more of the AI compute stack. McKinsey released a new report yesterday, estimating that by 2030, inference will account for ~60% of AI compute demand, versus ~40% for training. That makes inference optimization increasingly consequential: every gain in throughput and utilization changes how much useful AI the same hardware can serve. McKinsey’s full report ↓ mckinsey.com/industries/semiconductors/our-insi…
Linkmckinsey.commckinsey.com/industries/semiconductors/our-insights/scaling-ai-inferenceValidator hotkey 5GP7c3fFazW9GXK8Up3qgu2DJBk8inu4aK9TZy3RuoSWVCMi announced new child keys on SN14, SN118, applying at block 9,228,353 (~24h). Stake delegated to it there routes through its children after that. On SN14 it replaces the change that was pending for block 9,227,406, which will not apply then.
Between blocks 9,221,107 and 9,221,166
Validator hotkey 5GP7c3fFazW9GXK8Up3qgu2DJBk8inu4aK9TZy3RuoSWVCMi announced new child keys on SN14, applying at block 9,227,406 (~24h). Stake delegated to it there routes through its children after that. On SN14 it replaces the change that was pending for block 9,225,834, which will not apply then.
Between blocks 9,220,148 and 9,220,208View subnet →
Validator hotkey 5GP7c3fFazW9GXK8Up3qgu2DJBk8inu4aK9TZy3RuoSWVCMi announced new child keys on SN14, applying at block 9,225,834 (~24h). Stake delegated to it there routes through its children after that.
Between blocks 9,218,561 and 9,218,663View subnet →
Validator hotkey 5Deo9Q3SnZWNPSEPRTgapbyy6egKu2oaNSqD4RQHHJEuzD9Y announced new child keys on SN4, SN5, SN6, SN8, SN9, SN10, SN13, SN14, SN16, SN18, SN20, SN21 and 55 more, applying at block 9,212,198 (~24h). Stake delegated to it there routes through its children after that.
Between blocks 9,204,985 and 9,205,045
Validator hotkey 5CsvRJXuR955WojnGMdok1hbhffZyB4N5ocrv82f3p5A2zVp announced new child keys on SN6, SN14, applying at block 9,206,530 (~24h). Stake delegated to it there routes through its children after that.
Between blocks 9,199,286 and 9,199,348
The subnet burns 0% of its miner emission now (was 100%): miners are being paid again.
Detected 14d agoView subnet →
Validator hotkey 5GP7c3fFazW9GXK8Up3qgu2DJBk8inu4aK9TZy3RuoSWVCMi announced new child keys on SN14, SN120, applying at block 9,205,003 (~24h). Stake delegated to it there routes through its children after that. On SN14 it replaces the change that was pending for block 9,200,924, which will not apply then.
Between blocks 9,197,752 and 9,197,814
The subnet now routes 100% of its miner emission to burn (was 0%). Miners earn little or nothing from it while this lasts.
Detected 14d agoView subnet →
Validator hotkey 5CqRMuYSSqYkSa8AM3VawVfdyUhyBufyesvvgh3vCgaGb9Ed announced new child keys on SN14, SN92, SN93, SN111, applying at block 9,201,344 (~24h). Stake delegated to it there routes through its children after that.
Between blocks 9,194,135 and 9,194,197
Validator hotkey 5GP7c3fFazW9GXK8Up3qgu2DJBk8inu4aK9TZy3RuoSWVCMi announced new child keys on SN14, applying at block 9,200,924 (~24h). Stake delegated to it there routes through its children after that.
Between blocks 9,193,690 and 9,193,752View subnet →
Validator hotkey 5DjtmV76CRvxZqd68fDwWyTmsfqUZ5EUXVJ21FKTBzwLT4Kd announced new child keys on SN9, SN12, SN13, SN14, SN15, SN18, SN20, applying at block 9,178,399 (~24h). Stake delegated to it there routes through its children after that.
Between blocks 9,171,190 and 9,171,253
Cacheon@cacheon_aiGreat to see the work we’ve been doing with @affine_io get a moment on stage at @exploitsummit. Here’s @const_reborn telling the story in his opening keynote ↓
Validator hotkey 5DjtmV76CRvxZqd68fDwWyTmsfqUZ5EUXVJ21FKTBzwLT4Kd announced new child keys on SN14, SN17, SN18, SN30, SN31, SN37, SN38, SN51, SN54, applying at block 9,166,508 (~24h). Stake delegated to it there routes through its children after that.
Between blocks 9,159,305 and 9,159,371
Validator hotkey 5Deo9Q3SnZWNPSEPRTgapbyy6egKu2oaNSqD4RQHHJEuzD9Y announced new child keys on SN5, SN6, SN8, SN9, SN10, SN13, SN14, SN15, SN17, SN19, SN20, SN21 and 63 more, applying at block 9,162,013 (~24h). Stake delegated to it there routes through its children after that.
Between blocks 9,154,806 and 9,154,869
Cacheon@cacheon_aiQwen3.6 arena is now open for submissions! Better yet, we've generalized the evaluation system to support arbitrary kernel changes. No more waiting on slot updates to submit your winning kernel! We're excited to see what you can do. Submit today and earn Alpha right away.
Cacheon@cacheon_aiGLM-5.3 hits a new milestone on Cacheon: ~12.5% above its earlier benchmark, measured under different test conditions. On SN14 in $TAO Bittensor, miners optimize GPU operations to accelerate inference. We test changes with the full model running, and accepted work can support further improvements. Our latest competition submission reached 1,242.8 tokens per second. We're contributing to raise the bar and challenge other miners to push GLM-5.3 further. Our team follows the same evaluation, submission-order and acceptance rules as every other miner. We'll continue contributing as long as our involvement helps advance the subnet. Any emissions we earn will fund development and operations. In parallel, we're working to turn these advances into commercial opportunities, including collaborations with other subnets. We'll share more soon!
Cacheon@cacheon_aiIn our last competition, Cacheon miners pushed MiniMax M3 throughput to 2,337.9 tok/s: +32.2% vs the SGLang baseline. What might that kind of performance improvement be worth commercially? +32.2% throughput = ~61% more profit. How? SemiAnalysis InferenceX™ models a B300 running MiniMax M3 on vLLM at $4.81 profit per chip-hour. Apply the same +32.2% throughput gain, with pricing and chip costs held constant, and modeled profit rises to $7.73 per chip-hour. Why the difference? The GPU still costs $4.25 for the hour. More throughput means more capacity to sell without buying another chip-hour. That's a scenario, not measured Cacheon profit. What is measured is the optimization: 2,337.9 tok/s, +32.2%, produced by Bittensor miners and passed through Cacheon's testing and reproduction. Check the baseline. Run your own numbers.
