Never came across this issue on Model 3 before... Tesla Model 3 windshield wiper fluid spray stopped working Had front window replaced Wiper fluid was as good as empty Refilled wiper fluid Spray audibly working, but…. No washy washyshow more

Had to perform some more maintenance on my Tesla... Model Y today. It’s been nuts this winter. I spent a total of $10 on some wiper fluid. 😜show more

As I was working on my 2022 recap, I... came across this timelapse video I captured but never shared. It’s probably the smoothest timelapse of Jupiter’s rotation I’ve ever done, captured with an 11” telescope.show more

Breaking news 🗞️ 🚨 Tesla just quietly solved a... problem in Australia. The Model X is gone from our market. But the new Model Y L Premium AWD might actually be the closest thing we have to a replacement. And honestly… it makes a lot of sense. ⚡ Tesla Model Y L – Key Specs • 0–100 km/h: ~5.0 sec • Range: ~681 km WLTP • Top speed: 201 km/h • Seating: 6 adults • Supercharging: 250 kW • ~288 km added in 15 min 💰 Australian pricing (before on-road costs) Model Y Long Range AWD → $68,900 Model Y L Premium AWD → ~$74,900 So for roughly $6k more, you get: ✔ 3 rows ✔ 6 seats ✔ Much larger cabin ✔ ~400L extra cargo capacity ✔ Longer wheelbase ✔ Even more range 📊 Quick comparison Model Y Long Range AWD • 5 seats • ~600 km range • 0–100 km/h: 4.8 sec • $68,900 Model Y L Premium AWD • 6 seats • ~681 km range • 0–100 km/h: 5.0 sec • ~$74,900 So performance drops slightly, but practicality goes way up. 🇦🇺 Why this matters in Australia Since Tesla stopped selling the Model X locally, there has been a real gap in the lineup for larger families. The Model Y L doesn’t completely replace the X. You lose things like: ❌ Adaptive air suspension ❌ Driver instrument cluster ❌ Falcon Wing doors ❌ Some luxury interior touches But you still get: ✔ Tesla software ecosystem ✔ Supercharger network ✔ Massive range ✔ Practical 3-row seating And at a much lower price than a Model X ever was. 👨‍👩‍👧‍👦 Who this is perfect for • Growing families • Current Model Y owners needing more space • Former Model X buyers • Anyone considering EV9 / EX90 but wanting Tesla’s ecosystem Personally, as a Model X owner, this is the first Tesla sold in Australia that actually feels like a realistic successor. I’m seriously considering replacing my Model X with the Model Y L, possibly around the end of Q2 or mid-Q3 this year. Not a perfect Model X replacement. But for Australia right now? This might be Tesla’s smartest family vehicle yet. ⚡🇦🇺 ORDER NOW : Tesla Australia & New Zealand Tesla AIshow more

My friend who bought a new Model 3 today... was convinced that the 3 couldn’t be quieter than his Genesis. He had an injury that basically made his voice the volume of a whisper years ago Talking in the phone in the car is trouble for him because he can’t speak louder than a quiet whisper. I told him to try voice command in the Model 3 and the look in his face as the car picked up every word he said precisely was so fun to see the surprise on his face! He called me on his way home saying this is the best most quietest car he’s ever driven. Before today he has never driven any tesla and now 8 hours later can’t enjoy driving anything else!🔥🔥🔥show more

This art piece titled “Can’t Help Myself” is programmed... to try to contain the hydraulic fluid that’s constantly leaking out and required to keep itself running… if too much escapes it will die, so it’s desperately trying to pull it back to continue to fight for another day. saddest part is they gave the robot the ability to do ‘happy dances’ for spectators while the spill was contained. When the project was first launched in 2016, it danced around spending most of its time interacting with the crowd with exuberance, since it could quickly pull back the small spillage. many years later it looks worn down and hopeless, because the amount of leaked fluid became unmanageable as the spill grew over time. There now isn’t enough time to dance, as it only has enough time to try to keep it self alive. living its last days in a never-ending cycle between sustaining life and simultaneously bleeding out. the arm slowly came to a halt and died in 2019, but with a twist; the robot actually runs off of electricity, not hydraulics, so it was working its entire life towards some thing it didn’t even need, tricked by the system it was brought into. Wild how this parallels the human experience. can’t help but wonder if there’s a lesson to be learned from this art, mechanical or not. Does the suffering ever end? (From Aninarts on insta)show more

Our aero kit for the 2024+ Model 3 Performance... is designed using CFD to enhance performance while preserving Tesla’s renowned efficiency. Shown here is the exploded view of the aero kit parts to show how we accomplish our goal: to work with the Model 3 Performance foundation without disturbing it. The kit now delivers 1896N of downforce, shifting the car from net lift to substantial downforce with a minimal drag increase of just 1.93%. This is achieved through advanced aerodynamic techniques, such as splitter tunnels that enhance front ground effect—features rarely seen on street-legal vehicles. The kit enhances downforce in areas that already have high-pressure air, such as above the grille, over the side inlets, and on top of the front splitter, but also introduces entirely new shapes like the front undertray. Airflow management around the wheels and side skirts is critical for reducing drag and offsetting the added drag from downforce-generating surfaces. Additionally, placing a wing that is highly optimized to minimally disturb Tesla’s refined wake is crucial for maintaining efficiency. While we’re working out a few more final touches with our production run, we are just as excited to reach out to those who have patiently waited and look forward to providing order updates soon!show more

Moonshot AI is casually giving developers free daily access... to Kimi K3 😳 no subscription no upfront payment just sign in and start using one of the largest open AI models available what you get for $0: - Kimi K3 with 2.8T parameters - 1M token context window - strong coding and reasoning performance - native vision capabilities - free daily credits that refresh automatically why this is worth checking: > access a frontier model without paying API fees > long context for large codebases and documents > works on web, desktop, mobile, and CLI getting started takes less than 2 minutes: 1. go to 2. create a free account 3. Kimi K3 is available as the default model 4. start chatting or coding with your daily free credits bonus: Moonshot Together lets you invite friends for a chance to earn 3, 7, 15, 30, or even 365 days of Kimi Membership through its rewards program benchmark highlights: > 2.8T parameter MoE model > 1M context window > strong performance across coding, browsing, and reasoning benchmarks important: free credits reset daily, rate limits apply on the free tier, and the open-weight release is expected on July 27 A simple way to try one of the latest frontier AI models without paying for API accessshow more

The genial sanctions! For those who said that factories... like the Chinese ones exist in the West, no, they don't. The Chinese are using full-cycle automation in factories that even lack basic lighting. And this is working for cars as well as for missiles, as I have shown here before. Now comes the most interesting part. A large portion of the automation machinery is still European, but this dependence has been decreasing, and today it’s less than 40%. The West used to dominate the market for Chinese chips, industrial machinery, aviation, and various other sectors. What was the big idea? Impose sanctions on the Chinese and force them to develop all these technologies so they no longer need to buy from the West. Genius, don’t you think? Today, the Chinese are developing four commercial aviation engines simultaneously, with at least one already flying in testing phase, and three other engine projects for cargo aviation based on Russian models. This was a market that was entirely owned by the West. Another issue that will need to be addressed sooner or later is whether there will be tariffs on products coming from fully roboticized factories. Some might say this is punishing efficiency. But what’s the point of efficiency if no one has a job and can spend? An extremely efficient model that creates unemployment beyond its borders? Obviously, this will be discussed at some point.show more

I have chosen to drive my Tesla on FSD... multiple times instead of flying. In the past year, I’ve done multiple ~20-hour days driving continuously on FSD, covering over 1,000 miles daily without ever needing to touch the steering wheel. We just took a trip from Toronto to Nashville as well. It was 900 miles one way, which we completed in a single day (18 hours), and it was way more economical (and fun) than flying. Before that, I did Toronto to Austin, TX (3,200 miles round trip) TWICE in July and November — all in my Model 3 Performance. Would I attempt these drives without FSD? No way. I literally push a button on the screen, and the car handles all the charging stops, traffic, navigation, and everything. It’s so easy. If you’re an avid road tripper, there’s no better car to do it in than a Tesla.show more

I was on the tail end of a 500... mile drive back from Omaha yesterday with my 77 year old dad and we nearly had a very serious accident, possibly deadly for the other driver. Look at this older Toyota Corolla blow this light at 60mph more than 3 seconds after it turns red. If it was me driving, because I could not see the Corolla as it was occluded by the stopped cars, we would have had a head on collision or I'm mid-turn and he would have impacted my dad's side of my Cybertruck. Thankfully Tesla FSD was doing the driving and it has superhuman reaction to brake pedal application time (aka PRT = perception response time) of about 0.25s while very good human is 0.8-0.9s or average human is 1.5-2.5s. This is life saving technology and it may have saved one yesterday. Thanks to Ashok Elluswamy, Elon Musk, and the whole team behind this.show more

here's how the whole thing works. claude code doesn't... care what's behind the API. it just sends requests and expects responses. so i pointed it at my own machine instead of anthropic's servers. llama-server runs the model locally. LiteLLM sits in between and translates the API format. claude code thinks it's talking to claude. it's talking to qwen on localhost. the setup: 2x 3090s, 38 layers on GPU, 10 on CPU. 128K context window. generation is only 7 tok/s but the tradeoff is worth it. 128K means the agent can hold an entire project in memory without losing context midtask. claude code alone loads a 17.5K token system prompt on every request. tool definitions, safety rules, agent behavior. that's your baseline before you even say hello. pushed as far as i could tonight. what surprised me most wasn't the speed. it was the iteration quality. first prompt gave me a working particle sim. second prompt, the model read its own 564 lines, understood the architecture, and added trails, explosions, gravity wells, bloom effects. no handholding. 4bit quantized. 45GB on two consumer cards. running a full coding agent autonomously. detailed article coming. full benchmarks, hardware breakdowns, engine debugging, code quality. everything from setup to what broke and why.show more

HERMES AGENT NOW RUNS CLAUDE OPUS 5. NEAR FABLE... 5 INTELLIGENCE. HALF THE PRICE. SELF-VERIFIES ITS OWN WORK. AVAILABLE TODAY VIA NOUS PORTAL (20% OFF ALL MODELS). Anthropic shipped Opus 5 on July 24, 2026. same $5/$25 per million tokens as Opus 4.8. but the benchmarks tell a different story. WHAT CHANGED FROM OPUS 4.8: FrontierBench v0.1: Opus 5: 43.3%. Opus 4.8: 18.7%. 2.3x jump on the same test. ARC-AGI-3: Opus 5: 30.2%. 3x better than the next closest model. beat Fable 5 on 8 out of 13 benchmarks. at half the cost ($5/$25 vs $10/$50). same price as Opus 4.8. twice the intelligence. no reason to stay on 4.8. THE SPECS: model ID: claude-opus-5 context: 1M tokens (default and maximum) max output: 128K tokens thinking: on by default effort toggle: low / medium / high per request fast mode: $10/$50, 2.5x faster knowledge cutoff: May 2026 minimum cacheable prompt: 512 tokens (was 1,024) SELF-VERIFICATION (the biggest change): Opus 5 checks its own work automatically. Anthropic says: delete your verification prompts. "include a final verification step" now causes OVER-verification because the model already does it. for Hermes /goal tasks this is a direct upgrade. the judge checks evidence. the model also checks evidence. double layer of verification without extra tokens. EFFORT TOGGLE: low: fast, cheap, routine work. medium: balanced, daily tasks. high: full reasoning, complex problems. set per request. not a global switch. matches Hermes /reasoning command: /reasoning low (routine) /reasoning high (complex) Opus 5 effort toggle + Hermes reasoning control = precise cost management per turn. WHERE OPUS 5 FITS IN HERMES: DAILY DRIVER (replaces Opus 4.8): same price. 2.3x better benchmarks. set as your main model: Desktop app / Dashboard: Models → claude-opus-5 CHIEF OF STAFF: synthesis across multiple agents. reads Kanban, prioritizes, routes tasks. self-verification catches routing errors before they cascade. COMPLEX CODING: SOTA on agentic coding benchmarks. FrontierBench 43.3% = best public model for coding. set as coder profile model. /GOAL TASKS: self-verification + completion contracts = the model proves its work AND double-checks the proof. long-horizon goals finish correctly more often. MoA AGGREGATOR: strongest synthesis model at $5/$25. pair with GPT-5.6 and Grok 4.5 as references. Opus 5 aggregates. best quality at mid-range price. presets: max-quality: reference_models: - provider: openai-codex model: gpt-5.6-sol - provider: xai model: grok-4.5 aggregator: provider: anthropic model: claude-opus-5 COMPUTER USE: near-Fable 5 quality for browser automation. at half the token cost per session. computer_use tasks burn lots of vision tokens. Opus 5 halves that bill vs Fable 5. WHAT TO KEEP OPUS 5 AWAY FROM: cron monitoring: too expensive. use DeepSeek or no_agent mode. sub-agent grunt work: use GPT-5.6 Luna ($1/$6) or DeepSeek. auxiliary tasks: use Gemini Flash. routine web extraction: use a cheap model. Opus 5 is for the turns where quality compounds. planning, synthesis, verification, complex reasoning. budget models handle everything else. NOUS PORTAL: 20% OFF ALL MODELS Nous Portal currently runs a 20% discount on all models including Opus 5. $5/$25 official → $4/$20 through Nous Portal. the cheapest way to run Opus 5 right now. hermes setup --portal select claude-opus-5 as your model. discount applies automatically. Opus 5 replaces Opus 4.8 everywhere. same price. better at everything. no tradeoff. straight upgrade. hermes update /model claude-opus-5show more

"This was filmed last Wednesday afternoon at Riverside Veterinary... Clinic in Indianapolis, Indiana. The officer is Sergeant Paul Greer. He's 41 years old. Fourteen-year veteran of the Indianapolis Metropolitan Police Department. The dog is Bruno. A ten-year-old German Shepherd who served eight years as Paul's K9 partner before a joint condition ended his working career two years ago. When Bruno retired from active duty, Paul adopted him immediately. Brought him home. Bruno spent his retirement on Paul's couch, on Paul's bed, in the passenger seat of Paul's personal truck. The transition from working partner to household companion was seamless. Bruno had always been Paul's dog. The badge and the vest were just part of the job. Over the past several months, Bruno's condition had declined steadily. The joint condition spread. He had difficulty getting up. Stopped eating regularly. Paul had been managing Bruno's comfort with guidance from Dr. Angela Reese at Riverside for months. Last Tuesday evening, Bruno stopped getting up entirely. Paul called Dr. Reese that night. Wednesday afternoon, Paul drove Bruno to Riverside. He carried Bruno in from the truck himself. Wouldn't let the techs take him. Paul's partner, Officer Dana Choi, came with him. She filmed quietly on her phone from the corner of the room. She told us afterward that she asked Paul's permission before she started recording. He nodded. Paul sat on the exam table with Bruno cradled across his lap and chest. Bruno's head rested against Paul's shoulder. His eyes were half-open. His breathing was slow and easy. Paul bowed his head and pressed his face into Bruno's fur. Bruno lay still for a long moment. Then slowly — carefully — he raised both front paws. One at a time. And wrapped them around Paul's shoulders. And held on. Paul made a sound that Dana said she will never forget. Dr. Reese, who was standing nearby preparing, went completely still. Her assistant took a step back. Nobody moved. Dana told us: 'Bruno could barely lift his head that morning. But he lifted his paws and he held Paul. In that moment, with everything he had left, he held him. I think he was saying thank you. I think he was saying goodb"show more

UPDATE: Charlie Kirk 🚨 Muzzle Flash: Second Shooter Location,... Reflection, or Both? This video captures a flash on a window right as Charlie Kirk was shot. Let’s break this down… follow the numbers on the videos. #1. This video shows what appears to be a muzzle flash, just as Charlie is shot, leading people to believe the shooter was to Charlie’s right.. Notate the people to the left of the flash #2. This photo shows the location of the flash. Notate people to the left, on a middle-landing on a staircase. #3. This photo shows a clearer image showing both the middle-landing and the same location of the flash. It is a window. #4 . This video shows the building behind Charlie, which is a long hallway. This debunks any claims the shot came from here. — I zoom in where Charlie was — I zoom in on staircase — I zoom in on the window/flash — I zoom in on where suspected shooter was #5. I notate the flash reflection angles. — Video angle #1 notated — two reflection paths notated in relation to Video angle #1. 1. Reflection angle shows where we were told the shooter was. 2. Reflection angle shows mirrored angle, obviously where there is no reports of a shooter being in. 🔻 Final conclusion: This to me is very likely a muzzle flash reflection from a far off location. I am unsure of the reflection angle however, but this can be 100% proven if someone were to recreate the angles… maybe with a flash camera. If anyone is willing to, or has the means to do this (safely), this will either prove the shot came where the FBI said it came from, or it will prove there was a second shooter, in the direction of the mirrored angle.show more

I hope this post gets more than my average... 60 views bc it’s important. Please repost if so inclined: In the clip below from today’s Dan Bongino montage of Comey, the corrupt former FBI Director who caused chaos with the Russia Collusion Hoax, he unilaterally cleared HRC for mishandling classified materials and destroying evidence before the 2016 election. During this Press Conference I was at the FBI office with my agents working on a cartel prosecution. We all stopped to watch with amazement as Comey was initially eviscerating HRC’s unlawful conduct, but as we all now know he eventually stated “no reasonable prosecutor would bring such a case!” I remember turning towards my agents and saying “well, I guess I’m an unreasonable Prosecutor . . .” I knew in that moment at my deepest core, based on years of prosecutorial experience and instinct in federal law enforcement, that the reason Comey cleared HRC had nothing to do with the prosecutability of that fact pattern - it was political interference in its most corrupt form. And here’s why. Comey had NO authority to decline prosecution. At the end of the day, he was “just a cop” so to speak. His ONLY authority was to turn the investigation over to the DOJ which has sole discretion on who to prosecute, or not. He can make a recommendation, but not a decision to not prosecute, especially based on the reason stated. That’s why this will go down as one of the most embarrassing and demoralizing moments in FBI history. My agents were crushed. They knew what just happened as much as I did. This man stepped out of his lane to cover and protect a politician, while at the same time, without any of us knowing in that moment, he was also beginning the years long political assasination of Trump in what later became the Russia Hoax.show more

THE TESLA MODEL S: THE CAR THAT MADE ELECTRIC... VEHICLES SERIOUS When the Model S launched in 2012, the entire world still saw EVs as slow, boring, short-range toys for tree-huggers. The Model S changed that narrative overnight. It wasn’t just an electric car — it was a statement. Here’s why the Model S was so important for EV adoption: • It proved EVs could be faster and better than gas cars 0–60 mph in under 4 seconds (later Plaid versions under 2 seconds) while being completely silent and smooth. It beat most supercars off the line and made “electric” synonymous with performance. • It delivered real long-range capability Over 300 miles of range when most EVs at the time struggled to reach 100 miles. Suddenly, road trips became possible and “range anxiety” started to feel outdated. • It introduced over-the-air updates The first production car that could get major performance upgrades, new features, and safety improvements wirelessly — like a smartphone on wheels. This changed how people think about car ownership forever. • It forced the entire auto industry to respond Legacy manufacturers who had been dragging their feet on EVs suddenly rushed to catch up. The Model S basically lit the fuse for the modern EV revolution. • It made luxury electric desirable Premium interior, massive touchscreen, ridiculous acceleration, and futuristic design turned EVs from “compromise” into “aspiration.” Without the Model S proving that electric cars could outperform and out-luxury gasoline vehicles, we wouldn’t have the Model 3/Y explosion, the Cybertruck, or the flood of competitors now racing to go electric. The Model S didn’t just sell cars. It changed the future of transportation. It took EVs from niche to mainstream and showed the world what was possible.show more

This is my "feel the AGI" moment: I used... GPT-5.6 Sol to train my own autocorrect model that outperforms GPT-5.6 Sol (wtf??) I have no ML background. I have no idea what I'm doing. I just kept pushing Sol until it spat out a SOTA model. And I spent $0. The motivation: Years of talking to AI have made me terrible at typing. Rather than fix my skill issue, I decided to throw more AI at it. My idea was: instead of autocorrect that interrupts my flow, I want to type fast with mistakes and have AI clean it up after. I wanted the smallest local model possible, for speed, for battery life, for science! So I decided to train my own. Inspired by Andrej Karpathy’s autoresearch, I ran Codex /goal with this setup: pick an experiment, try it, record the results to a doc, throw it out if it fails, and plan the next experiment without repeating failures. I gave a few examples that had to pass, tight latency targets, and let it run. Sol did some amazing things. First, it scanned benchmarks and shortlisted base models: Qwen 3.5, Gemma 4, Liquid LFM 2.5. It found a dataset on HuggingFace for typed text. Then it built a simulator for fingers striking a Mac keyboard, modeling the physical layout with a Gaussian distribution around each key. It simulated striking the wrong key, wrong order, fat-fingering, etc. With the models + data + simulator, it fine-tuned using MLX right on my MacBook. It had a working prototype within an hour! But accuracy was pretty poor. — Problem 1: Tokenization Sol read papers, ran tests, and identified that the tokenizer was the bottleneck. Tokenization makes typos hard for the model to see, so it memorizes mappings instead of using its language priors. Sol tried ByT5, Google’s tokenizer-free byte-level LLM. This made a big improvement, but the model is old and lacked the knowledge needed to reach Sol performance. Sol dug deeper and realized a tokenizer-free model isn’t needed; instead, it used T5Gemma, an encoder-decoder model. This can understand the input deeply before producing output, and furthermore, Sol could post-train the encoder to improve performance. This gave a much higher ceiling. — Problem 2: Loss function Now the model was correcting some typos perfectly, but ignoring most. Sol realized that standard cross-entropy loss was teaching the model to avoid edits, because the vast majority of characters in the training data were left unmodified. The fix was wild: Sol wrote a custom loss function that byte-aligns the source and target strings, uses a dynamic programming algorithm to compute the minimum edits between the two, then weights correct edits much higher than copies. After a lot of tuning, this dramatically improved accuracy. — Problem 3: Autoregression One failure mode remained: if the model made a mistake, it couldn’t backtrack. It could only predict the next token. Teaching it to “think” like a reasoning model would solve this, but would be far too slow. Sol found a beautiful solution: instead of greedily predicting the next token, beam search over all possibilities. This parallelizes the exploration instead of one linear chain-of-thought. At the end, choose the path with highest cumulative log probability. This worked great, but made the experience worse, since the user wouldn’t see progress until the whole search was done. To fix this, Sol made a clever observation: after each search step, the longest common prefix among surviving branches is guaranteed to appear in the final result, so it can be displayed immediately. As the search progresses, weaker paths are dropped and the prefix grows, so the user sees continuous progress. Sol built all this as a custom MLX pipeline that does the parallel decoding on the MacBook GPU, with just ~40ms TTFT. It’s crazy fast and entirely local. — Final eval (error reduction rate, higher is better): - Apple autocorrect: 49.66% - GPT-5.6 Luna: 82.47% - GPT-5.6 Terra: 87.64% - GPT-5.6 Sol: 90.56% - Our model (1.7B): 91.02% Final cost: - 1 quota reset (thanks Tibo) - $0 (And yes, I verified there's no cheating. In fact, we test words scrubbed from the training data to prove the model isn’t memorizing) There were a ton more details and tangents I could write about: contrastive learning, GRPO, DPO, dynamic masking, and more. Sol is a fascinating and creative model. It blew my mind so many times. Don’t let a lack of experience stop you: Sol makes AI experiments accessible to anyone!show more

my 8 GB VRAM gaming laptop is absolutely going... to hate me for this. but I still did it. ran a 31b dense model (Gemma 4 31b Q4) with only 8 GB VRAM last week I ran Gemma 4 26B A4B a mixture of experts model on my RTX 4060 and hit 25–28 tokens/sec using llama.cpp's new MTP support. smooth. snappy. but MoE has a secret: it only activates 4B parameters per token despite having 26B total. that's why it flies. so the real question started haunting me. what if I throw a full, no tricks, every parameter fires on every token, 31B DENSE model at the same machine? # Hardware: GPU: NVIDIA RTX 4060, 8 GB VRAM RAM: 16 GB CPU: Intel Core i7 H Laptop. Gaming. Modest. The model: gemma-4-31B-it-qat-UD-Q4_K_XL.gguf (model's unsloth huggingface link in the comments) This is Google DeepMind's flagship dense model in the Gemma 4 family that can run on single consumer GPU. It packs a hybrid attention architecture, supports up to 256K context natively, and is QAT (Quantization Aware Training) optimized, meaning it retains far more quality than standard post training quants at the same bit depth. This is NOT the MoE. This is 31 BILLION dense parameters, every single one of them loaded. # the flags I used: -m gemma-4-31B-it-qat-UD-Q4_K_XL.gguf -cnv --spec-type draft-mtp --spec-draft-model mtp-gemma-4-31B-it.gguf --spec-draft-n-max 8 --spec-draft-p-min 0.6 -c 6000 -v Multi Token Prediction (MTP) is still active here. Separate draft GGUF required, same as the 26B setup. # Results: → Decode: ~3 tokens/sec → Prefill: ~2 tokens/sec → Context: 6000 tokens → Hardware crying quietly in the corner: yes so is 3 tps actually usable? For real time back and forth chat? Not ideal. You're not having a fluid conversation at 3 tps. but slow ≠ useless. And this is where it gets genuinely interesting. think about how senior devs actually work in a real team. But when something is architectural, deeply complex, or needs serious reasoning? they walk down the hall and escalate to the senior. That's exactly the local AI agent architecture this unlocks: → Fast orchestrator model (Gemma 4 26B MoE at 25+ tps) handles routing, simple queries, tool calls, memory. The junior dev. → Gemma 4 31B dense is the senior, called only when the fast model genuinely hits a wall. Hard multi step reasoning. Complex code generation. Deep architectural decisions. The agentic loop stays fast. Only the hard hops touch the 31B. That's a legitimate production grade local AI architecture on a budget hardware. (requires 2 8gb gpus) other workflows where 3 tps is completely fine: - overnight batch jobs. summarize documents, extract structured data, review code. Fire it off. Sleep. wake up to results. - One shot deep reasoning - Silent code audit loops, you write and test, the 31B reviews diffs and flags issues in the background between your sprints - Any workflow where output quality > output speed A few weeks ago, nobody was running a 30B+ dense model on a single consumer GPU with 8 GB VRAM. At all. Now we're doing it on an Intel i7-H gaming laptop with a NVIDIA RTX 4060, thanks to llama.cpp + QAT quants + MTP speculative drafting. Google DeepMind said the Gemma 4 31B targets "consumer GPUs and workstations." They were not exaggerating. The hardware bar to run serious frontier class models locally keeps dropping. the tools are here. the models are here. you just have to be willing to abuse your laptop a little. what workflows would you actually run on a local 3 tps 31B dense model? genuinely curious. drop it below.show more

Out of ewc. What a perfect way to go... out as well. Thinking that i have 0 idea how to play fighting games LOL. Everything i learned my entire life was thrown out the window in that losers match. And with that, time to never turn on fatal fury ever again!!! Right before ENC imma play the game for a couple hours, but thats it. The game patched me out of existence. SNK just doesnt like my style of play, and honestly, im ok with that. Im glad at evo japan I was able to show everyone how good I am, and im glad i was able to qualify for this event. No regrets. Goodluck to everyone still in the tournament!!! To the supporters, sorry I wasnt good enough ='(show more