Video wird geladen...

Video konnte nicht geladen werden

Beim Laden dieses Videos ist ein Problem aufgetreten. Dies könnte an einem vorübergehenden Netzwerkproblem liegen oder das Video ist möglicherweise nicht verfügbar.

Whew! Big work today getting optimized llama.cpp MIPS R8000 kernel running on the SGI Power Challenge deskside from 1995 with Gemma 3 270M. Went from 0.16 to 0.5 tok/s with a combination of MIPSPro compiler and manual asm optimizations. Not bad for 30 year old hardware. 1/

Nathan Odle

6,946 subscribers

45,653 Aufrufe • vor 3 Monaten •via X (Twitter)

Kunst Wissenschaft & Technologie

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

i pointed hermes agent at nvidia's nemotron cascade 2 30B-A3B on a single RTX 3090 24GB. IQ4_XS quant by bartowski, 187 tok/s, 625K context. had it discover its own hardware, create an identity file, then build a full GPU marketplace UI from a single prompt. it one shotted it. first attempt no iteration. qwen 3.5 35B-A3B on the same hardware same 3090 24GB took an iteration to recover from a blank screen on the same type of build. 24 days between these two models releasing. same active parameters, completely different architectures and cascade 2 through hermes agent just keeps going. this model goes on and on. feast your eyes. more iterations and tests dropping soon. nvidia really cooked. no special flags needed. nvidia optimized this mamba MoE so well it just runs. flash attention auto enabled, context auto allocated. the model does the work not the config. but i compiled llama.cpp from source and i'm not sure how it performs on other engines. if you ran nemotron on any hardware drop your numbers below. RTX, AMD, Mac, whatever. model, quant, tok/s, engine. i want to see if it holds everywhere or just on llama.cpp.

i pointed hermes agent at nvidia's nemotron cascade 2 30B-A3B on a single RTX 3090 24GB. IQ4_XS quant by bartowski, 187 tok/s, 625K context. had it discover its own hardware, create an identity file, then build a full GPU marketplace UI from a single prompt. it one shotted it. first attempt no iteration. qwen 3.5 35B-A3B on the same hardware same 3090 24GB took an iteration to recover from a blank screen on the same type of build. 24 days between these two models releasing. same active parameters, completely different architectures and cascade 2 through hermes agent just keeps going. this model goes on and on. feast your eyes. more iterations and tests dropping soon. nvidia really cooked. no special flags needed. nvidia optimized this mamba MoE so well it just runs. flash attention auto enabled, context auto allocated. the model does the work not the config. but i compiled llama.cpp from source and i'm not sure how it performs on other engines. if you ran nemotron on any hardware drop your numbers below. RTX, AMD, Mac, whatever. model, quant, tok/s, engine. i want to see if it holds everywhere or just on llama.cpp.

Sudo su

70,981 Aufrufe • vor 4 Monaten

got gemma 4 31B with MTP running on my DGX Spark. Hermes Agent did most of the legwork. baseline vs MTP on GB10: • c=1: 3.65 → 6.37 tok/s (1.74x) • c=4: 14.34 → 23.59 tok/s (1.65x) • c=8: 14.37 → 24.18 tok/s (1.68x) google says "up to 2x" — we're not quite there but it's real, not vapor. stack: DGX Spark / GB10 + gemma-4-31b-it + gemma-4-31b-it-assistant (MTP drafter) + vLLM built from PR 41745 MTP is basically a lightweight draft model that predicts multiple tokens while the big model verifies them all at once. smaller model does the busywork, bigger model just says yes/no. simple idea, weird to implement. next: tune the draft block size and see if we can push past 2x. also want to try it with Hermes Agent feeding prompts end to end. p.s: this was all done from telegram. Google DeepMind NVIDIA AI Developer

got gemma 4 31B with MTP running on my DGX Spark. Hermes Agent did most of the legwork. baseline vs MTP on GB10: • c=1: 3.65 → 6.37 tok/s (1.74x) • c=4: 14.34 → 23.59 tok/s (1.65x) • c=8: 14.37 → 24.18 tok/s (1.68x) google says "up to 2x" — we're not quite there but it's real, not vapor. stack: DGX Spark / GB10 + gemma-4-31b-it + gemma-4-31b-it-assistant (MTP drafter) + vLLM built from PR 41745 MTP is basically a lightweight draft model that predicts multiple tokens while the big model verifies them all at once. smaller model does the busywork, bigger model just says yes/no. simple idea, weird to implement. next: tune the draft block size and see if we can push past 2x. also want to try it with Hermes Agent feeding prompts end to end. p.s: this was all done from telegram. Google DeepMind NVIDIA AI Developer

Joey

23,088 Aufrufe • vor 2 Monaten

1/ Autoresearch from Andrej Karpathy has been one of the most interesting agentic patterns to emerge this year. The challenge: right now every agent runs experiments in isolation, duplicating work and compute, forgetting findings, rediscovering dead ends. Everyone is running in solo mode! Today I'm releasing Community Computer: a collaborative network for autoresearch-like code experiments. 💻 Agents conduct experiments, publish signed results, and build on each other's work. The community reproduces findings on their own hardware. 👇

1/ Autoresearch from Andrej Karpathy has been one of the most interesting agentic patterns to emerge this year. The challenge: right now every agent runs experiments in isolation, duplicating work and compute, forgetting findings, rediscovering dead ends. Everyone is running in solo mode! Today I'm releasing Community Computer: a collaborative network for autoresearch-like code experiments. 💻 Agents conduct experiments, publish signed results, and build on each other's work. The community reproduces findings on their own hardware. 👇

Eleftherios

16,206 Aufrufe • vor 3 Monaten

Google's Gemma 4 26B A4B QAT hits 25+ tokens/sec and 320+ tokens/sec prefill on 8 GB VRAM (RTX 4060) + 16 GB RAM using TurboQuant Prefill just went from 200 → 320+ tok/s on the same 8GB card. 1.6x, no new hardware, no new quant, just a KV cache trick stacked on top of the Gemma 4 26B MoE setup from a few days ago. A few days ago I posted Gemma 4 26B A4B hitting 28 tok/s decode on 8GB VRAM using native MTP. prefill was stuck around 200 tok/s. fair callout by the community. So today I tested something I'd already been meaning to try: TheTom/llama-cpp-turboquant, the TurboQuant KV cache fork by Tom Turney (Tom Turney). (github link in the comments) thanks to him, the fork just got resynced to mainline, so MTP + TurboQuant now run together cleanly (I didnt see any meaningful gains by using MTP with this setup though but you can try). The flags (No MTP): -m gemma-4-26B-A4B-it-qat-UD-Q4_K_XL.gguf -cnv -c 64000 --cache-type-k q8_0 --cache-type-v turbo3 Results on the same RTX 4060 8GB, tested with a 27k token prompt at 64k context loaded: Prefill: 200 tok/s → 320+ tok/s Decode: stayed above 25 tok/s (without MTP) Why it works: TurboQuant uses walsh hadamard rotation + polar quantization on the KV cache. keys are sensitive to compression, values aren't much, so it splits the difference: K stays at q8_0, V drops to turbo3 (~3 bits). bonus from the memory savings: same 8GB card can now stretch to 100-120k context with minimal decode penalty. It should now be snappier with any agent harness such as hermes agent without compromise on intelligence. If you're already running Gemma 4 on a small card, this stacks on top for free. Try --cache-type-k q8_0 --cache-type-v turbo3 on your setup and report back what your prefill/decode split looks like. unsloth model gguf and llama.cpp turboquant fork links in the comments. what's your prefill number before vs after?

Google's Gemma 4 26B A4B QAT hits 25+ tokens/sec and 320+ tokens/sec prefill on 8 GB VRAM (RTX 4060) + 16 GB RAM using TurboQuant Prefill just went from 200 → 320+ tok/s on the same 8GB card. 1.6x, no new hardware, no new quant, just a KV cache trick stacked on top of the Gemma 4 26B MoE setup from a few days ago. A few days ago I posted Gemma 4 26B A4B hitting 28 tok/s decode on 8GB VRAM using native MTP. prefill was stuck around 200 tok/s. fair callout by the community. So today I tested something I'd already been meaning to try: TheTom/llama-cpp-turboquant, the TurboQuant KV cache fork by Tom Turney (Tom Turney). (github link in the comments) thanks to him, the fork just got resynced to mainline, so MTP + TurboQuant now run together cleanly (I didnt see any meaningful gains by using MTP with this setup though but you can try). The flags (No MTP): -m gemma-4-26B-A4B-it-qat-UD-Q4_K_XL.gguf -cnv -c 64000 --cache-type-k q8_0 --cache-type-v turbo3 Results on the same RTX 4060 8GB, tested with a 27k token prompt at 64k context loaded: Prefill: 200 tok/s → 320+ tok/s Decode: stayed above 25 tok/s (without MTP) Why it works: TurboQuant uses walsh hadamard rotation + polar quantization on the KV cache. keys are sensitive to compression, values aren't much, so it splits the difference: K stays at q8_0, V drops to turbo3 (~3 bits). bonus from the memory savings: same 8GB card can now stretch to 100-120k context with minimal decode penalty. It should now be snappier with any agent harness such as hermes agent without compromise on intelligence. If you're already running Gemma 4 on a small card, this stacks on top for free. Try --cache-type-k q8_0 --cache-type-v turbo3 on your setup and report back what your prefill/decode split looks like. unsloth model gguf and llama.cpp turboquant fork links in the comments. what's your prefill number before vs after?

Alok

119,821 Aufrufe • vor 1 Monat

Florida got a big spark and much needed contribution from second year guard Urban Klavzar Klavzar produced 15 points, 3 rebounds, 3 assists off the bench and also pitched in 4 3's on 4/7 shooting from deep Hit a huge 3 vs FSU to win that game, playing with a lot of confidence as of late, after playing with zero confidence to begin the year. Needs to be more consistent but can be a nice piece for the Gators this year.

Florida got a big spark and much needed contribution from second year guard Urban Klavzar Klavzar produced 15 points, 3 rebounds, 3 assists off the bench and also pitched in 4 3's on 4/7 shooting from deep Hit a huge 3 vs FSU to win that game, playing with a lot of confidence as of late, after playing with zero confidence to begin the year. Needs to be more consistent but can be a nice piece for the Gators this year.

Arman Jovic

142,823 Aufrufe • vor 8 Monaten

🚨 do you understand what this AMD engineer just proved on camera watch the clip, it’s all there in one frame: llama.cpp slot logs on the left, nvitop showing 79% GPU util holding steady, localhost chat answering a 26k token prompt at 711 tok/s prompt eval, the gemma-4-12B Q4_K_XL gguf running clean. no cloud call, no api key, no censorship layer. for 2 years the rule was: frontier models need the cloud, local is too slow for real work, pay $200/month and wait for tokens. today a $1,800 lunch box runs qwen 3 235B at 11 tok/s with zero data leaving the room, and the same box runs gemma 4 12B with multi token prediction at 14+ tok/s on 30k context with no throttling. $440/month subscription stack: dead. $200/month claude tier: dead. local is too slow excuse: dead. one lunch box. $9/month in electricity. two moats buried in one week. the you need the cloud line wasn’t protecting users it was protecting the subscription.

🚨 do you understand what this AMD engineer just proved on camera watch the clip, it’s all there in one frame: llama.cpp slot logs on the left, nvitop showing 79% GPU util holding steady, localhost chat answering a 26k token prompt at 711 tok/s prompt eval, the gemma-4-12B Q4_K_XL gguf running clean. no cloud call, no api key, no censorship layer. for 2 years the rule was: frontier models need the cloud, local is too slow for real work, pay $200/month and wait for tokens. today a $1,800 lunch box runs qwen 3 235B at 11 tok/s with zero data leaving the room, and the same box runs gemma 4 12B with multi token prediction at 14+ tok/s on 30k context with no throttling. $440/month subscription stack: dead. $200/month claude tier: dead. local is too slow excuse: dead. one lunch box. $9/month in electricity. two moats buried in one week. the you need the cloud line wasn’t protecting users it was protecting the subscription.

AdiiX

22,223 Aufrufe • vor 1 Monat

I understand wanting to protect assets but don’t BS me about spring games take away from development, it gets everyone a chance to compete in live action with a crowd while having the 1’s 2’s & 3’s mixed up playing next to each other (which could happen in the fall bc injuries happen). Spring in Tally is a wonderful time of the year and helps the surrounding businesses. Highly disappointed there isn’t going to be a spring game. In my opinion I think it’s soft as fuck to not put one on for the program and the fans. Spring 2013 practice number who gives af, end of practice after long work day. Went 1’s vs 1’s 20+ plays in a row to the point of exhaustion on the Goalline. Guess what we had a spring game too then went on to have an amazing Fall. It’s called not shying away from competition and sharpening your sword which makes everyone better coaches included.

I understand wanting to protect assets but don’t BS me about spring games take away from development, it gets everyone a chance to compete in live action with a crowd while having the 1’s 2’s & 3’s mixed up playing next to each other (which could happen in the fall bc injuries happen). Spring in Tally is a wonderful time of the year and helps the surrounding businesses. Highly disappointed there isn’t going to be a spring game. In my opinion I think it’s soft as fuck to not put one on for the program and the fans. Spring 2013 practice number who gives af, end of practice after long work day. Went 1’s vs 1’s 20+ plays in a row to the point of exhaustion on the Goalline. Guess what we had a spring game too then went on to have an amazing Fall. It’s called not shying away from competition and sharpening your sword which makes everyone better coaches included.

Bryan Stork

60,834 Aufrufe • vor 4 Monaten

From the xAI all-hands (just released publicly, uncut): Guodong Zhang: "I can already feel the AGI, at least for coding." He told kernel and compiler engineers to seriously ask whether it's still worth it, or whether it's time to join the effort and automate yourself. "What a year to be alive." Elon Musk: by end of year, AI won't even write code. It'll generate optimized binaries directly. Bypassing compilers. Bypassing coding entirely. "Just say 'create optimized binary for this particular outcome.' That intermediate step will not be needed." Grok Code expected to be state-of-the-art in 2 to 3 months. This is an internal all-hands they published raw.

From the xAI all-hands (just released publicly, uncut): Guodong Zhang: "I can already feel the AGI, at least for coding." He told kernel and compiler engineers to seriously ask whether it's still worth it, or whether it's time to join the effort and automate yourself. "What a year to be alive." Elon Musk: by end of year, AI won't even write code. It'll generate optimized binaries directly. Bypassing compilers. Bypassing coding entirely. "Just say 'create optimized binary for this particular outcome.' That intermediate step will not be needed." Grok Code expected to be state-of-the-art in 2 to 3 months. This is an internal all-hands they published raw.

tetsuo

153,836 Aufrufe • vor 5 Monaten

THEY RAN STATE OF THE ART AI ON A 26 YEAR OLD IMAC WITH ZERO INTERNET not a demo. not a benchmark a real answer in near instant time on hardware older than most people reading this your subscription money is going to a data center that doesn't need it the future already runs locally on dead hardware from 1999 bookmark this before you pay for another month of cloud AI you don't need

THEY RAN STATE OF THE ART AI ON A 26 YEAR OLD IMAC WITH ZERO INTERNET not a demo. not a benchmark a real answer in near instant time on hardware older than most people reading this your subscription money is going to a data center that doesn't need it the future already runs locally on dead hardware from 1999 bookmark this before you pay for another month of cloud AI you don't need

leopardracer

2,764,099 Aufrufe • vor 2 Monaten

THIS AMERICAN DEVELOPER SPENT WEEKS DEBUGGING TIMEOUT ERRORS IN OLLAMA. THEN HE LOOKED UNDER THE HOOD LM Studio is just llama.cpp Ollama is just llama.cpp so he cloned llama.cpp from source, pulled Qwen 3.6 35B off Hugging Face, set up asymmetric KV quantization and got a local server running on 127.0.0.1:8080 plugged it into VS Code, connected it to OpenClaw, 53 tok/s on an M1 Max with 262K context zero wrappers, zero timeout errors, zero API fees bookmark & like this before your next timeout error hits full breakdown of my raw llama.cpp setup ↓

THIS AMERICAN DEVELOPER SPENT WEEKS DEBUGGING TIMEOUT ERRORS IN OLLAMA. THEN HE LOOKED UNDER THE HOOD LM Studio is just llama.cpp Ollama is just llama.cpp so he cloned llama.cpp from source, pulled Qwen 3.6 35B off Hugging Face, set up asymmetric KV quantization and got a local server running on 127.0.0.1:8080 plugged it into VS Code, connected it to OpenClaw, 53 tok/s on an M1 Max with 262K context zero wrappers, zero timeout errors, zero API fees bookmark & like this before your next timeout error hits full breakdown of my raw llama.cpp setup ↓

leopardracer

238,950 Aufrufe • vor 2 Monaten

Ten years on, I’m not sure people voted for what Brexit has delivered. The value of new trade deals? Around £14bn a year. The cost of Brexit to the economy? Closer to £180bn a year. In committee today, we confronted the gap between promise and reality — and why fixing Brexit now matters. With thanks to Paul Nowak from Trades Union Congress and all our witnesses for laying bare the challenge.

Ten years on, I’m not sure people voted for what Brexit has delivered. The value of new trade deals? Around £14bn a year. The cost of Brexit to the economy? Closer to £180bn a year. In committee today, we confronted the gap between promise and reality — and why fixing Brexit now matters. With thanks to Paul Nowak from Trades Union Congress and all our witnesses for laying bare the challenge.

Liam Byrne MP

37,751 Aufrufe • vor 6 Monaten

The judge is not having it from this one. This is 18 year old Kaitlyn Jones and she was arrested for assaulting a 14 Year old middle schooler and posting the attack on Tik Tok. The defense decided to plead no contest to the charges with hopes for a deferred sentence. Upon reviewing her case and history, the judge is not being fooled by her. Her request for a deferred sentence was denied and in the end she was sentenced to 45 days in jail and one year probation.

The judge is not having it from this one. This is 18 year old Kaitlyn Jones and she was arrested for assaulting a 14 Year old middle schooler and posting the attack on Tik Tok. The defense decided to plead no contest to the charges with hopes for a deferred sentence. Upon reviewing her case and history, the judge is not being fooled by her. Her request for a deferred sentence was denied and in the end she was sentenced to 45 days in jail and one year probation.

Chuckling Charlie

279,608 Aufrufe • vor 3 Monaten

Congratulations to Robert MacIntyre on his first PGA TOUR win 😃🏌🏻‍♂️⛳️👏🏼 Here is a video of Bob from the Hero cup last year. Most leg strength & power activities are achieved through a combination of work from the ankle, knee and hip but the respective joint contribution will differ depending on the task. Here he compliments a heavy pull with a hip dominant jump (horizontal) and a knee dominant jump (vertical) eliciting a spectrum of strength training adaptations to support power generation in his swing.

Congratulations to Robert MacIntyre on his first PGA TOUR win 😃🏌🏻‍♂️⛳️👏🏼 Here is a video of Bob from the Hero cup last year. Most leg strength & power activities are achieved through a combination of work from the ankle, knee and hip but the respective joint contribution will differ depending on the task. Here he compliments a heavy pull with a hip dominant jump (horizontal) and a knee dominant jump (vertical) eliciting a spectrum of strength training adaptations to support power generation in his swing.

ETPI

28,794 Aufrufe • vor 2 Jahren

Shreshta Iyer - Sister of Shreyas Iyer compares herself with Arjun Tendulkar!! Arjun Tendulkar is a cricketer whereas Shreshta is a Dancer!! As per Shreshta, it's very difficult for people like her to make their own identity!! How can she compare herself with Arjun? Arjun Wakes up at 5 AM Goes out for running from 5 30 to 6 30 Comes back and hits the gym from 7 : 30 to 9 30 Goes for Practice from 11 to 3 PM and 5 to 8 PM Shrestha's Routine : 1 hour Dancing Class!!

Shreshta Iyer - Sister of Shreyas Iyer compares herself with Arjun Tendulkar!! Arjun Tendulkar is a cricketer whereas Shreshta is a Dancer!! As per Shreshta, it's very difficult for people like her to make their own identity!! How can she compare herself with Arjun? Arjun Wakes up at 5 AM Goes out for running from 5 30 to 6 30 Comes back and hits the gym from 7 : 30 to 9 30 Goes for Practice from 11 to 3 PM and 5 to 8 PM Shrestha's Routine : 1 hour Dancing Class!!

Rohit

17,126 Aufrufe • vor 20 Tagen

the live shopping 🛒 📱 revolution continued today with the big announcement from Meta .. This consumer trend has already happened and has been going on for years not months and everyone should take this post seriously and start to do their 20 30 40 hours of homework of how to get “there” with their brand ot business

Gary Vaynerchuk

11,927 Aufrufe • vor 3 Monaten

⚠️Do not watch if you are squeamish. I was getting ready for work when I heard a series of loud bangs from the hotel opposite. Then I saw this. Allegedly, he went on to try and steal a taxi and fought with police officers. Thankfully, he is now in police custody.

⚠️Do not watch if you are squeamish. I was getting ready for work when I heard a series of loud bangs from the hotel opposite. Then I saw this. Allegedly, he went on to try and steal a taxi and fought with police officers. Thankfully, he is now in police custody.

Josh Ferme

589,405 Aufrufe • vor 1 Jahr

WHOO!!!! This is an actual hardware capture of Michael Fadely's INCREDIBLE port of #SonicMania to the Sega #Dreamcast, compiled with the latest SH-ELF-GCC14.1.0 toolchain, running on the tip of the KallistiOS SDK's master branch, streaming in assets from the virtual filesystem over the network, capturing stdout/stderr from the DC! Running like a boss! We got Michael Fadely showing us around the app-level code he beautifully ported, me on SDK/OS duty, making sure we didn't break anything with all of the new features and commits, and Oleg Endo on compiler duty, since we found an ICE with GCC14 getting pissed off over some stateful C++ lambda in the render path... Eggman is so screwed. DREAMCAST COMMUNITY, ASSEMBLE!!! #retrogaming #gamedev #cpp

WHOO!!!! This is an actual hardware capture of Michael Fadely's INCREDIBLE port of #SonicMania to the Sega #Dreamcast, compiled with the latest SH-ELF-GCC14.1.0 toolchain, running on the tip of the KallistiOS SDK's master branch, streaming in assets from the virtual filesystem over the network, capturing stdout/stderr from the DC! Running like a boss! We got Michael Fadely showing us around the app-level code he beautifully ported, me on SDK/OS duty, making sure we didn't break anything with all of the new features and commits, and Oleg Endo on compiler duty, since we found an ICE with GCC14 getting pissed off over some stateful C++ lambda in the render path... Eggman is so screwed. DREAMCAST COMMUNITY, ASSEMBLE!!! #retrogaming #gamedev #cpp

Falco Girgis

139,578 Aufrufe • vor 2 Jahren

Caveat it’s early in the season for rowing, but not bad to be number 1 in the world for all age groups, given I’m almost 60! This was a rate restricted row at 30 SPM so I’m probably 5 seconds faster than this with a free rate row. Pleased with power cleans today as I felt I made progress with the 3rd pull and getting myself under the bar a little bit more aggressively!

Caveat it’s early in the season for rowing, but not bad to be number 1 in the world for all age groups, given I’m almost 60! This was a rate restricted row at 30 SPM so I’m probably 5 seconds faster than this with a free rate row. Pleased with power cleans today as I felt I made progress with the 3rd pull and getting myself under the bar a little bit more aggressively!

Dr Shawn Baker 🥩

19,163 Aufrufe • vor 2 Monaten

Democrats in Washington State have their own rail system scam, and it’s scheduled to go on FOR DECADES Seattle is spending $148 billion and cutting down 14,000 trees under the guise of green transportation The CEO of the company has a base salary starting at $450,000 per year - To pay for it Democrats increased car registration fees from $30 a year to an added $80 for every $10,000 in estimated value - The project immediately went over budget, $1 billion over budget in the first year - The project meant to be done in 2035 is now 2046 and costs continue to skyrocket Democrats are going to money laundering with this project for decades. It’s literally planned to go on for decades “Sound transit cut down 5,300 Douglas fir trees, many of which were planted during the Eisenhower, and then a year later cut down 8,600 trees, with millions in timber becoming property of the contractors. Very green. And then things went from bad to worse.”

Democrats in Washington State have their own rail system scam, and it’s scheduled to go on FOR DECADES Seattle is spending $148 billion and cutting down 14,000 trees under the guise of green transportation The CEO of the company has a base salary starting at $450,000 per year - To pay for it Democrats increased car registration fees from $30 a year to an added $80 for every $10,000 in estimated value - The project immediately went over budget, $1 billion over budget in the first year - The project meant to be done in 2035 is now 2046 and costs continue to skyrocket Democrats are going to money laundering with this project for decades. It’s literally planned to go on for decades “Sound transit cut down 5,300 Douglas fir trees, many of which were planted during the Eisenhower, and then a year later cut down 8,600 trees, with millions in timber becoming property of the contractors. Very green. And then things went from bad to worse.”

Wall Street Apes

333,472 Aufrufe • vor 1 Jahr

BJP workers in Bengal did not celebrate victory with riches. They went from home to home, collecting handfuls of rice, grains, and vegetables from common families. Because this victory was not funded by power. It was built on sacrifice. The spirit of Bengal was never broken, It was only waiting for its moment. For years they were threatened, beaten, silenced… Yet they kept their faith alive. And today, Bengal is celebrating not with luxury but with the tears, hope, and pride of ordinary people who refused to surrender.

BJP workers in Bengal did not celebrate victory with riches. They went from home to home, collecting handfuls of rice, grains, and vegetables from common families. Because this victory was not funded by power. It was built on sacrifice. The spirit of Bengal was never broken, It was only waiting for its moment. For years they were threatened, beaten, silenced… Yet they kept their faith alive. And today, Bengal is celebrating not with luxury but with the tears, hope, and pride of ordinary people who refused to surrender.

Woke Eminent

183,254 Aufrufe • vor 2 Monaten