Video yükleniyor...

Video Yüklenemedi

Bu video yüklenirken bir sorun oluştu. Bu geçici bir ağ sorunundan kaynaklanıyor olabilir veya video kullanılamıyor olabilir.

Ana Sayfaya Dön

Here's a comparison i never thought i would make Qwen-3.8-27b vs Gemini-3.7-flash Yes, you read it right! I don't know what to say, this is soo damn impressive. the difference is hugeee🤯 Take a look at this result 😍 (watch it in highest quality) Same prompt, both one shot... show more

AJ

8,170 subscribers

302,905 görüntüleme • 1 ay önce •via X (Twitter)

Bilim & Teknoloji Komedi Eğitim

Anya Rossi• Live Now

Private livecam show

33 Yorum

AJ profil fotoğrafı

AJ1 ay önce

prompt: Create a Three.js water simulation. Requirements: Plane mesh with 512x512 subdivisions Wave propagation simulation Mouse click creates ripples Multiple interacting waves Real-time lighting Everything in a single HTML file No external libraries besides Three.js

AJ profil fotoğrafı

AJ1 ay önce

y'all supernova cooked with this one. What a model 😍

Leon Cvetkovski profil fotoğrafı

Leon Cvetkovski1 ay önce

considering how cheap 3.7 flash is its probably a MoE around 60-70b, so actually a fair comparison

AJ profil fotoğrafı

AJ1 ay önce

Overall this is a good win for 27b.

Need More VRAM profil fotoğrafı

Need More VRAM1 ay önce

Gemini Flash 3.7 dead on arrival

AJ profil fotoğrafı

AJ1 ay önce

So happy with 3.8-27b now

Jason from BallotScore.com profil fotoğrafı

Jason from BallotScore.com1 ay önce

I was getting 38-45t/s out of unsloth's Qwen3.8-27b earlier today consistently out of a single 3090 and 64GB DDR4 RAM.

AJ profil fotoğrafı

AJ1 ay önce

With MTP right? im switching between Q4/Q5 as i see fit.

Pardox profil fotoğrafı

Pardox1 ay önce

Man you can make YouTube vidoes but more in depth?

AJ profil fotoğrafı

AJ1 ay önce

I can, anything in depth you like to see here, i can prob make a post faster.

0xBakeer profil fotoğrafı

0xBakeer1 ay önce

It is a great win for qwen 🚀

Ajay profil fotoğrafı

Ajay1 ay önce

Comparing anything with Gemini models is a crime. Gemini is garbage through and through.

AJ profil fotoğrafı

AJ28 gün önce

Haha valid reaction lol

Clara Lafever Jane profil fotoğrafı

Clara Lafever Jane1 ay önce

Wow 👌 👏 😍

Ddddddd profil fotoğrafı

Ddddddd1 ay önce

Q5 is significantly better than q4?

Zion Sati profil fotoğrafı

Zion Sati1 ay önce

What harness are you using?

AJ profil fotoğrafı

AJ1 ay önce

For that demo, just from webchat ui The harness i use most is pi, then Hermes and opencode

Draneil Mifa profil fotoğrafı

Draneil Mifa1 ay önce

On 3090 on Q5 ? What about your context length ? How do you do ? I ve same 3090, can you share your llama.cpp settings ? 🙏

HLEET3D profil fotoğrafı

HLEET3D1 ay önce

Very impressive, do you think Qwen 3.8-27B can handle controlling other LLM ? just like opus can do

SourceCodeplz profil fotoğrafı

SourceCodeplz1 ay önce

wow local is real now again! wish coding subs would have them

Daniil Volkov profil fotoğrafı

Daniil Volkov29 gün önce

Hi! Big thanks for gguf! Do your models have MTP?

Fayaz Mohammad profil fotoğrafı

Fayaz Mohammad1 ay önce

Which is which, it's not clearly visible

AJ profil fotoğrafı

AJ1 ay önce

It's in header, left is qwen, right is gemini

Ella Tech & Tool profil fotoğrafı

Ella Tech & Tool1 ay önce

That Qwen result on a 3090 is wild 🤯 Local winning

NITHIN STEVEN F profil fotoğrafı

NITHIN STEVEN F1 ay önce

What's average tps here ?

Doğukan profil fotoğrafı

Doğukan1 ay önce

cool results! I’ll try this on 2bit, really curious about how it will handle

notableelbaton profil fotoğrafı

notableelbaton1 ay önce

RTX3090 24GB, with how much RAM, and which CPU, and is that vLLM or llama.cpp ?

Henry Fan profil fotoğrafı

Henry Fan1 ay önce

Amazing! Just a technical question: how did you manage the context window? I can't load a long one and it have to do multiple compact even with just one prompt.

Deepakraj Solanki profil fotoğrafı

Deepakraj Solanki1 ay önce

But what should be the laptop specification

AsT3RoiD profil fotoğrafı

AsT3RoiD1 ay önce

@GeminiApp hello

Lonelytrader profil fotoğrafı

Lonelytrader1 ay önce

wait,why my qwen 3.8 27b make this for me? q8 model

AJ profil fotoğrafı

AJ1 ay önce

Try with this prompt "Create a Three.js water simulation. Requirements: Plane mesh with 512x512 subdivisions Wave propagation simulation Mouse click creates ripples Multiple interacting waves Real-time lighting Everything in a single HTML file No external libraries besides Three.js "

ShipCircuit profil fotoğrafı

ShipCircuit1 ay önce

Incredible model it is. Im really impressed

Benzer Videolar

Gemini 3.8 Flash vs Opus 5 same prompt to each model and i wasn't expecting gemini 3.8 flash to be this fast! build time: > Opus 5: 24 minutes > 3.8 Flash: 37 seconds 💀 if gemini is given same time as opus 5 it will mog the hell out of opus in terms of quality too which one would you choose here?

Gemini 3.8 Flash vs Opus 5 same prompt to each model and i wasn't expecting gemini 3.8 flash to be this fast! build time: > Opus 5: 24 minutes > 3.8 Flash: 37 seconds 💀 if gemini is given same time as opus 5 it will mog the hell out of opus in terms of quality too which one would you choose here?

Tim Jayas

76,990 görüntüleme • 29 gün önce

Alright guys, wow. Qwen 3.8 27B is basically Qwen 4. What we have here is one of the best one-shot shark survival games I've made with any model. I ran this same prompt through fable, and the biggest thing all models struggle with in this test is the top-down view of the sharks. Fable even had all of the fins reversed and details jumbled on the first turn. Before now, it was really a 2 to 8 turn test to get something final regardless of model. Qwen 3.8 27B just nailed this in one output, no harness at all, just purely the model thoroughly thinking before outputting the final result. On every asset used, different shark variants, everything, it all looks incredibly tight. The boat movement, shark speed and engagement, everything is so much better than I expected, and I was expecting a lot. Imagine what 8 turns would do in a harness? This confirms to me that while Qwen 3.8 27B thinks a lot, for the first time in a local model on the base weights, IT MIGHT ACTUALLY BE WORTH IT This is a Frontier Lab model result, and it ran on my 5090 at 60tps in Q5_K_M. I am just in complete astoundment at what Qwen has accomplished here, and I am incredibly grateful that this model is open source. The speed will also increase drastically with optimisations. This is just an incredible day. What a blessing. More to come!

Alright guys, wow. Qwen 3.8 27B is basically Qwen 4. What we have here is one of the best one-shot shark survival games I've made with any model. I ran this same prompt through fable, and the biggest thing all models struggle with in this test is the top-down view of the sharks. Fable even had all of the fins reversed and details jumbled on the first turn. Before now, it was really a 2 to 8 turn test to get something final regardless of model. Qwen 3.8 27B just nailed this in one output, no harness at all, just purely the model thoroughly thinking before outputting the final result. On every asset used, different shark variants, everything, it all looks incredibly tight. The boat movement, shark speed and engagement, everything is so much better than I expected, and I was expecting a lot. Imagine what 8 turns would do in a harness? This confirms to me that while Qwen 3.8 27B thinks a lot, for the first time in a local model on the base weights, IT MIGHT ACTUALLY BE WORTH IT This is a Frontier Lab model result, and it ran on my 5090 at 60tps in Q5_K_M. I am just in complete astoundment at what Qwen has accomplished here, and I am incredibly grateful that this model is open source. The speed will also increase drastically with optimisations. This is just an incredible day. What a blessing. More to come!

Kyle Hessling

24,917 görüntüleme • 1 ay önce

Every big MoE model I test on my 3090 is basically to see how good it is compared to Qwen3.6-27B. "How good is it compared to Qwen3.6-27B?" is the question I get asked the most, and that shows how much people love and, more importantly, trust the 27B 27B is battle tested. It has proven its worth and has nothing left to prove. Now it just needs to make way for the next 27B. Here’s an example video: a Three.js task, same prompt, not oneshot but two shots for all models. All ran on my 3090 + 64GB VRAM Qwen-3.6-27b fully on vram while others have experts offloaded to system ram. Pinned on the left: Qwen3.6-27B-Q5 On the right, i'm switching between Inkling-Small-Q4_K_M DSV4-Flash-0731-UD-IQ1 Ling-3.0-Flash-Q4_K_M Qwen3.6-27B nailed it. The effects and visuals are just spot on.The other models got some things right, but not everything. Second place would have to go to Inkling-Small while DSV4-Flash-0731 and Ling-3.0-Flash failed to capture the full requirements. This is also why we’re all excited for Qwen Qwen3.8-27B!

Every big MoE model I test on my 3090 is basically to see how good it is compared to Qwen3.6-27B. "How good is it compared to Qwen3.6-27B?" is the question I get asked the most, and that shows how much people love and, more importantly, trust the 27B 27B is battle tested. It has proven its worth and has nothing left to prove. Now it just needs to make way for the next 27B. Here’s an example video: a Three.js task, same prompt, not oneshot but two shots for all models. All ran on my 3090 + 64GB VRAM Qwen-3.6-27b fully on vram while others have experts offloaded to system ram. Pinned on the left: Qwen3.6-27B-Q5 On the right, i'm switching between Inkling-Small-Q4_K_M DSV4-Flash-0731-UD-IQ1 Ling-3.0-Flash-Q4_K_M Qwen3.6-27B nailed it. The effects and visuals are just spot on.The other models got some things right, but not everything. Second place would have to go to Inkling-Small while DSV4-Flash-0731 and Ling-3.0-Flash failed to capture the full requirements. This is also why we’re all excited for Qwen Qwen3.8-27B!

AJ

11,670 görüntüleme • 1 ay önce

I dare you to recreate this!! Made by Gemini 3.8 Flash.. Just told it that make me a beachfront scenery path with mountains on the other side made only with random strokes of lines. this is so abstract and honestly impressive Gemini 3.8 Flash is goated in making abstract art!

I dare you to recreate this!! Made by Gemini 3.8 Flash.. Just told it that make me a beachfront scenery path with mountains on the other side made only with random strokes of lines. this is so abstract and honestly impressive Gemini 3.8 Flash is goated in making abstract art!

Bee

15,354 görüntüleme • 26 gün önce

glm 5.3 vs qwen 3.8 vs gemini 3.7 vs deepseek v4 flash four models designed and built three structures each on a physics-backed site, with no dimensions anywhere in the brief the setup: our own agent loop on OpenRouter, a construction site as the tool set – footings, walls, arches, roofs, scaffold, a lamp. the site enforces physics and nothing else: unsupported brick falls, a roof needs walls under it, a worker reaches 3.2 m above whatever he stands on, an arch needs centring until the keystone is set, concrete cures before it carries. no budget ceiling – material cost is tallied and reported, never blocked. tasks: 1. house – a plot and a palette, no plan. shape, height and material are the model's call 2. lighthouse – a headland cut by a gully, with a rock stack standing 30 m offshore. the lamp must burn, it must be the highest thing built, and the keeper must be able to walk to it 3. bridge – a river with one islet and banks at different heights. cross it however you want models: Z.ai glm 5.3 flash, Qwen qwen 3.8 flash, Google DeepMind gemini 3.7 flash, DeepSeek v4 flash vision all twelve objects were finished and signed off by the models themselves. tallest lighthouse is qwen's at 38.4 m, planted on the offshore stack with a bridge run out to it – the only model that read the site that way. deepseek signed off its bridge on an empty riverbed: 0 bricks, 107 minutes, $1.16m of material tallied - total cost, three builds #1 glm 5.3 flash – $0.201 #2 gemini 3.7 flash – $0.871 #3 qwen 3.8 flash – $1.058 #4 deepseek v4 flash – $1.567 - wall clock, three builds #1 gemini 3.7 flash – 91m #2 glm 5.3 flash – 228m #3 deepseek v4 flash – 502m #4 qwen 3.8 flash – 912m - total tokens #1 gemini 3.7 flash – 3,567,052 #2 glm 5.3 flash – 4,732,748 #3 qwen 3.8 flash – 13,469,333 #4 deepseek v4 flash – 18,230,076 - defects logged by the site #1 deepseek v4 flash – 59 #2 gemini 3.7 flash – 132 #3 glm 5.3 flash – 221 #4 qwen 3.8 flash – 350 - material tallied across three builds #1 gemini 3.7 flash – $359,884 #2 glm 5.3 flash – $583,358 #3 deepseek v4 flash – $1,327,484 #4 qwen 3.8 flash – $2,188,625 observations: • glm is the cheap one and nothing here is close – $0.201 for three buildings, $0.042 per million tokens, 6x under gemini's rate • what glm spends it on is bulk, not care: 166,228 bricks in one house and 156 defect weight, the worst single object in the set • gemini is the efficiency line – 91 minutes and 3.57m tokens for all three and an eighth of qwen's clock • gemini also builds the smallest of everything. its lighthouse is 22.5 m against qwen's 38.4, its house 6.9 m against 19.3 • qwen is the maximalist: 1.18m bricks, $2.19m of material, tallest on all three tasks, and 912 minutes – 15 hours – to get there conclusion: twelve finished objects for $3.80 all in, and a 7.8x price spread between the cheapest model and the priciest! follow thehype. for 24/7 ai news, analysis and breakdowns

glm 5.3 vs qwen 3.8 vs gemini 3.7 vs deepseek v4 flash four models designed and built three structures each on a physics-backed site, with no dimensions anywhere in the brief the setup: our own agent loop on OpenRouter, a construction site as the tool set – footings, walls, arches, roofs, scaffold, a lamp. the site enforces physics and nothing else: unsupported brick falls, a roof needs walls under it, a worker reaches 3.2 m above whatever he stands on, an arch needs centring until the keystone is set, concrete cures before it carries. no budget ceiling – material cost is tallied and reported, never blocked. tasks: 1. house – a plot and a palette, no plan. shape, height and material are the model's call 2. lighthouse – a headland cut by a gully, with a rock stack standing 30 m offshore. the lamp must burn, it must be the highest thing built, and the keeper must be able to walk to it 3. bridge – a river with one islet and banks at different heights. cross it however you want models: Z.ai glm 5.3 flash, Qwen qwen 3.8 flash, Google DeepMind gemini 3.7 flash, DeepSeek v4 flash vision all twelve objects were finished and signed off by the models themselves. tallest lighthouse is qwen's at 38.4 m, planted on the offshore stack with a bridge run out to it – the only model that read the site that way. deepseek signed off its bridge on an empty riverbed: 0 bricks, 107 minutes, $1.16m of material tallied - total cost, three builds #1 glm 5.3 flash – $0.201 #2 gemini 3.7 flash – $0.871 #3 qwen 3.8 flash – $1.058 #4 deepseek v4 flash – $1.567 - wall clock, three builds #1 gemini 3.7 flash – 91m #2 glm 5.3 flash – 228m #3 deepseek v4 flash – 502m #4 qwen 3.8 flash – 912m - total tokens #1 gemini 3.7 flash – 3,567,052 #2 glm 5.3 flash – 4,732,748 #3 qwen 3.8 flash – 13,469,333 #4 deepseek v4 flash – 18,230,076 - defects logged by the site #1 deepseek v4 flash – 59 #2 gemini 3.7 flash – 132 #3 glm 5.3 flash – 221 #4 qwen 3.8 flash – 350 - material tallied across three builds #1 gemini 3.7 flash – $359,884 #2 glm 5.3 flash – $583,358 #3 deepseek v4 flash – $1,327,484 #4 qwen 3.8 flash – $2,188,625 observations: • glm is the cheap one and nothing here is close – $0.201 for three buildings, $0.042 per million tokens, 6x under gemini's rate • what glm spends it on is bulk, not care: 166,228 bricks in one house and 156 defect weight, the worst single object in the set • gemini is the efficiency line – 91 minutes and 3.57m tokens for all three and an eighth of qwen's clock • gemini also builds the smallest of everything. its lighthouse is 22.5 m against qwen's 38.4, its house 6.9 m against 19.3 • qwen is the maximalist: 1.18m bricks, $2.19m of material, tallest on all three tasks, and 912 minutes – 15 hours – to get there conclusion: twelve finished objects for $3.80 all in, and a 7.8x price spread between the cheapest model and the priciest! follow thehype. for 24/7 ai news, analysis and breakdowns

thehype.

26,360 görüntüleme • 1 ay önce

This completely changes how I look at AI benchmarks. Qwen 3.8 Max Preview felt nearly unusable in OpenCode and Qwen Code. With the Claude Code harness, it suddenly came surprisingly close to Kimi K3 and Fable 5. Same model. Same task. Different harness. Wildly different result. How is this even possible?

This completely changes how I look at AI benchmarks. Qwen 3.8 Max Preview felt nearly unusable in OpenCode and Qwen Code. With the Claude Code harness, it suddenly came surprisingly close to Kimi K3 and Fable 5. Same model. Same task. Different harness. Wildly different result. How is this even possible?

OmedTheVibeCoder

30,223 görüntüleme • 2 ay önce

Qwen 3.8 27B beat Claude Opus 4.6 for $0 open-weights Qwen 3.8 27B on one rented GPU vs anthropic's Claude Opus 4.6. same prompt, no edits the test: one self-contained three.js file rendering sunken Atlantis — temple, columns, boid fish, kelp, bubbles, god rays, caustics — plus a 20-second camera flythrough cost: Qwen 3.8 27B — free (open weights, self-hosted) Claude Opus 4.6 — $0.21 what we saw: • qwen actually built a temple — full colonnade, domed roof, glowing portal centered. opus made a long row of columns fading into fog • qwen's portal is the star of the shot. opus leaves it small and off to the side • qwen packed more into frame — fish, kelp, ruins layered together. opus spreads thin and empty qwen's weights are on hugging face — you can run this one yourself!

Qwen 3.8 27B beat Claude Opus 4.6 for $0 open-weights Qwen 3.8 27B on one rented GPU vs anthropic's Claude Opus 4.6. same prompt, no edits the test: one self-contained three.js file rendering sunken Atlantis — temple, columns, boid fish, kelp, bubbles, god rays, caustics — plus a 20-second camera flythrough cost: Qwen 3.8 27B — free (open weights, self-hosted) Claude Opus 4.6 — $0.21 what we saw: • qwen actually built a temple — full colonnade, domed roof, glowing portal centered. opus made a long row of columns fading into fog • qwen's portal is the star of the shot. opus leaves it small and off to the side • qwen packed more into frame — fish, kelp, ruins layered together. opus spreads thin and empty qwen's weights are on hugging face — you can run this one yourself!

AI/ML API

102,960 görüntüleme • 1 ay önce

Yup! Gemini 3.8 Flash made this in Antigravity.. It's exceptionally good in frontend tasks.. i was amazed to see this level of polished output.. It did take a few iterations to tweak specific animations and morphing but this result is unreal! Which one do you like more?

Yup! Gemini 3.8 Flash made this in Antigravity.. It's exceptionally good in frontend tasks.. i was amazed to see this level of polished output.. It did take a few iterations to tweak specific animations and morphing but this result is unreal! Which one do you like more?

Bee

30,997 görüntüleme • 26 gün önce

🤯This is crazy... Qwen 3.8 Flash Next IS really good... it basically one shot this, then I just told it to add the MLX-Serve / Qwen structure after. (mixed 4-8bit version) (~90gb ram, it did use full 256k ctx) Prompt: build me a minecraft clone using threejs, npm, typescript, test files, linting, and better graphics than minecraft, make it look good. Just simple block building with the basic minecraft blocks, but make it look really good with amazing graphics. Generated textures and sounds. Benchmaxxing ? Maybe, needs more testing, but I have been seeing some good results on Youtube also.

🤯This is crazy... Qwen 3.8 Flash Next IS really good... it basically one shot this, then I just told it to add the MLX-Serve / Qwen structure after. (mixed 4-8bit version) (~90gb ram, it did use full 256k ctx) Prompt: build me a minecraft clone using threejs, npm, typescript, test files, linting, and better graphics than minecraft, make it look good. Just simple block building with the basic minecraft blocks, but make it look really good with amazing graphics. Generated textures and sounds. Benchmaxxing ? Maybe, needs more testing, but I have been seeing some good results on Youtube also.

David Dalcu

26,075 görüntüleme • 1 ay önce

this is what 12gb of vram builds in 2026, absolute magic > rtx 3060 12gb, #1 gpu on steam > bonsai 2 27b + mtp, 5.95 gb of weights > hermes agent, 5 hours, 328k tokens written > 8 js files, 2,368 lines, zero hand written code > 50 tok/s fresh, 22 tok/s average, 125k context watch the full video, 5 hours in 12 minutes of pure dance of a local ai model on rtx 3060 12gb vram, and stay till the end for the full gameplay. this entire game was built by bonsai2, a qwen 3.8 27b dense compressed to ternary, and this small model is punching way above its weight. it built multi file engineering work using hermes agent, sure it's not fast but perfect for overnights and routine work and the quality is insane, and context holding is another best one, it does not lose the thread. i ran PrismML bonsai 1 made from qwen 3.6 27b dense and i built things with it, but this time with the latest base model qwen 3.8 these results are insane, and because i loved building with it a lot i thought many more of you would run it because this gpu exists in almost every home. so i packaged mtp, doubled the speed from 26 tok/s to 50 tok/s, packed the prefill fix in and released it on huggingface, almost 4,000 downloads in 3 days. i'll leave a link below.

this is what 12gb of vram builds in 2026, absolute magic > rtx 3060 12gb, #1 gpu on steam > bonsai 2 27b + mtp, 5.95 gb of weights > hermes agent, 5 hours, 328k tokens written > 8 js files, 2,368 lines, zero hand written code > 50 tok/s fresh, 22 tok/s average, 125k context watch the full video, 5 hours in 12 minutes of pure dance of a local ai model on rtx 3060 12gb vram, and stay till the end for the full gameplay. this entire game was built by bonsai2, a qwen 3.8 27b dense compressed to ternary, and this small model is punching way above its weight. it built multi file engineering work using hermes agent, sure it's not fast but perfect for overnights and routine work and the quality is insane, and context holding is another best one, it does not lose the thread. i ran PrismML bonsai 1 made from qwen 3.6 27b dense and i built things with it, but this time with the latest base model qwen 3.8 these results are insane, and because i loved building with it a lot i thought many more of you would run it because this gpu exists in almost every home. so i packaged mtp, doubled the speed from 26 tok/s to 50 tok/s, packed the prefill fix in and released it on huggingface, almost 4,000 downloads in 3 days. i'll leave a link below.

Sudo su

76,833 görüntüleme • 8 gün önce

So this is the result of Gemini 4 Pro hiding under the name “Gemini 3.8 Flash” 🙄 I tried this so called rumored “Gemini 4 Pro” against famous "V8 Engine" demo from X and no, this is definitely not what all those posts about Gemini 4 Pro supposedly topping Arena benchmarks made it sound like. It does feel better than any Gemini model I’ve used so far, but it’s still not there. I wouldn’t even put it at Opus 5 level yet. Not sure what this hype is about, tbh.

So this is the result of Gemini 4 Pro hiding under the name “Gemini 3.8 Flash” 🙄 I tried this so called rumored “Gemini 4 Pro” against famous "V8 Engine" demo from X and no, this is definitely not what all those posts about Gemini 4 Pro supposedly topping Arena benchmarks made it sound like. It does feel better than any Gemini model I’ve used so far, but it’s still not there. I wouldn’t even put it at Opus 5 level yet. Not sure what this hype is about, tbh.

The Bugged Dev

28,496 görüntüleme • 14 gün önce

Qwen 3.8 27B vs Fable, Kimi K3, GPT 5.6 Sol Pro on Flappy Bird. Prompt: “Make the ultimate adorable cute and beautiful flappy bird game in HTML.” Qwen 3.8 27B Xhigh INT8: - Video 1 (Puff) - 8m 11s - 160 TPS (2x3090) Fable 5 Max: - Video 2 (Flappy Fluff) - 20m 21s Kimi K3 Max Fireworks Fast: - Video 3 (Fluffy Flap!) - 4m 40s - 111.3 TPS GPT 5.6 Sol Pro: - Video 4 (Flappy Boom) - 19m 7s Overview: Personally, I think Fable is the best. It feels extremely polished, sharp and stands out with small things like the dizzy effect on death. However, 27B is freakishly close. Never would I have guessed 2 months ago this would have come from a local model. 27B actually beats Fable on certain aspects like the 3D tilting ui buttons on death. Kimi K3 felt the best to play mechanics wise. It felt smooth with the perfect balance of difficulty and ease to play. Also, it was the only one that had music. 5.6 Sol Pro was hilariously bad in all aspects. I tried 5.6 xhigh multiple times and the outputs were even worse.

Qwen 3.8 27B vs Fable, Kimi K3, GPT 5.6 Sol Pro on Flappy Bird. Prompt: “Make the ultimate adorable cute and beautiful flappy bird game in HTML.” Qwen 3.8 27B Xhigh INT8: - Video 1 (Puff) - 8m 11s - 160 TPS (2x3090) Fable 5 Max: - Video 2 (Flappy Fluff) - 20m 21s Kimi K3 Max Fireworks Fast: - Video 3 (Fluffy Flap!) - 4m 40s - 111.3 TPS GPT 5.6 Sol Pro: - Video 4 (Flappy Boom) - 19m 7s Overview: Personally, I think Fable is the best. It feels extremely polished, sharp and stands out with small things like the dizzy effect on death. However, 27B is freakishly close. Never would I have guessed 2 months ago this would have come from a local model. 27B actually beats Fable on certain aspects like the 3D tilting ui buttons on death. Kimi K3 felt the best to play mechanics wise. It felt smooth with the perfect balance of difficulty and ease to play. Also, it was the only one that had music. 5.6 Sol Pro was hilariously bad in all aspects. I tried 5.6 xhigh multiple times and the outputs were even worse.

Youssof Al Toukhi

12,761 görüntüleme • 1 ay önce

Very quick comparison between Ornith 1.5 35B MoE and Qwen 3.8 27B Dense. 📣 Clearly it's not a fair one, but let's in any case see how it went. 397B download in progress! In the Videos below: - Brick SA -> Qwen 3.8 - Lego Streets -> Ornith 1.5 Context: - Pi agent used in both cases, prompt below - M3 Ultra 512GB - Ornith-1.5-35B-A3B-oQ8e MTP hosted on oMLX - mlx-community/Qwen3.8-27B-8bit DFlash 2 hosted on mlx-dspark - Ornith has been incredibly fast with speed from 75 t/s to 45 t/s (160K+ context) - Qwen3.8 suffered context much more reaching 5 t/s above 160K context, but it can be due to engine tested still work in progress - Prompt: using threejs, and cdn, create a lego like game that's inspired by gta san andreas, with beautiful aesthetics and graphics and ability to steal cars. it should be 3d and have a nice and large map and areas. the graphics should be decent and nice. it should feature iconic things from gta san andreas Notes: - Qwen 3.8 result is 0-shot, while Ornith 1.5 required 6 iterations - Ornith 1.5 is not at the same level of autonomy as Qwen 3.8 27B honestly. To try getting the same results I'm constantly nudging, steering and 🤬 at it. - Qwen 3.8 took 6 hours to complete, but it was more a problem of timeout of Chrome headless used for testing, real prefill/decode TBD. I'll test again now with oMLX - Pi agent has a nearly perfect Cache Hit ratio that for local models is MEGA important!

Very quick comparison between Ornith 1.5 35B MoE and Qwen 3.8 27B Dense. 📣 Clearly it's not a fair one, but let's in any case see how it went. 397B download in progress! In the Videos below: - Brick SA -> Qwen 3.8 - Lego Streets -> Ornith 1.5 Context: - Pi agent used in both cases, prompt below - M3 Ultra 512GB - Ornith-1.5-35B-A3B-oQ8e MTP hosted on oMLX - mlx-community/Qwen3.8-27B-8bit DFlash 2 hosted on mlx-dspark - Ornith has been incredibly fast with speed from 75 t/s to 45 t/s (160K+ context) - Qwen3.8 suffered context much more reaching 5 t/s above 160K context, but it can be due to engine tested still work in progress - Prompt: using threejs, and cdn, create a lego like game that's inspired by gta san andreas, with beautiful aesthetics and graphics and ability to steal cars. it should be 3d and have a nice and large map and areas. the graphics should be decent and nice. it should feature iconic things from gta san andreas Notes: - Qwen 3.8 result is 0-shot, while Ornith 1.5 required 6 iterations - Ornith 1.5 is not at the same level of autonomy as Qwen 3.8 27B honestly. To try getting the same results I'm constantly nudging, steering and 🤬 at it. - Qwen 3.8 took 6 hours to complete, but it was more a problem of timeout of Chrome headless used for testing, real prefill/decode TBD. I'll test again now with oMLX - Pi agent has a nearly perfect Cache Hit ratio that for local models is MEGA important!

Ivan Fioravanti ᯅ

12,150 görüntüleme • 1 ay önce

deepseek v4.1 flash ran all four three.js scenes for $0.15. gemini 3.8 flash needed $1.25 we put the two models on one job: a self-contained html page that renders an animated 3d scene with three.js – a storm at sea, a spiral galaxy, a tornado over farmland, a tesla coil on a dark table. same prompt - calls until all four scenes were usable #1 deepseek v4.1 flash – 6 #2 gemini 3.8 flash – 12 - time #1 deepseek v4.1 flash – 13m 57s #2 gemini 3.8 flash – 32m 02s - cost #1 deepseek v4.1 flash – $0.15 #2 gemini 3.8 flash – $1.25 observations: • gemini blew out the exposure on 4 of 5 first attempts – a white blob where the tesla coil should be. the prompt had a paragraph on exposure. it did not help. • both models overcorrect in the fix loop: "too bright" becomes "almost black". round two is where it lands. same four scenes, 6 calls and $0.15 against 12 calls and $1.25!

deepseek v4.1 flash ran all four three.js scenes for $0.15. gemini 3.8 flash needed $1.25 we put the two models on one job: a self-contained html page that renders an animated 3d scene with three.js – a storm at sea, a spiral galaxy, a tornado over farmland, a tesla coil on a dark table. same prompt - calls until all four scenes were usable #1 deepseek v4.1 flash – 6 #2 gemini 3.8 flash – 12 - time #1 deepseek v4.1 flash – 13m 57s #2 gemini 3.8 flash – 32m 02s - cost #1 deepseek v4.1 flash – $0.15 #2 gemini 3.8 flash – $1.25 observations: • gemini blew out the exposure on 4 of 5 first attempts – a white blob where the tesla coil should be. the prompt had a paragraph on exposure. it did not help. • both models overcorrect in the fix loop: "too bright" becomes "almost black". round two is where it lands. same four scenes, 6 calls and $0.15 against 12 calls and $1.25!

thehype.

62,603 görüntüleme • 22 gün önce

Qwopus 3.6-27b Coder I had a lot of requests to test it, so I did. I ran the same tests I’ve done on other models. It scored lower on tool-eval-bench than both Qwen 3.6 27b and Qwen 3.6 35b. It also produced broken single-file Tetris and Solar System HTML demos, even after 3 fresh attempts. In the Tetris html, the game stalled when you completed a line. It also randomly changed the pieces when moving them left or right. In the Solar System html, clicking on the sun or any planet broke the rendering — the objects just disappeared. Conclusion: The regular Qwen 3.6 27b and 35b models performed better. What do you think of Qwopus 3.6-27b Coder? Full results & RAW files: Full prompts in the posts below.

Qwopus 3.6-27b Coder I had a lot of requests to test it, so I did. I ran the same tests I’ve done on other models. It scored lower on tool-eval-bench than both Qwen 3.6 27b and Qwen 3.6 35b. It also produced broken single-file Tetris and Solar System HTML demos, even after 3 fresh attempts. In the Tetris html, the game stalled when you completed a line. It also randomly changed the pieces when moving them left or right. In the Solar System html, clicking on the sun or any planet broke the rendering — the objects just disappeared. Conclusion: The regular Qwen 3.6 27b and 35b models performed better. What do you think of Qwopus 3.6-27b Coder? Full results & RAW files: Full prompts in the posts below.

Mia

11,183 görüntüleme • 3 ay önce

I'm really happy with Qwen3.8-27B ! All frustrations of not able to use Ox-Alpha and running out of credits for Glm-5.3, Grok turned me into Qwen-3.8-27b. Running locally on my 3090. This is the first fps game i'm creating with Qwen3.8-27b. I was after NMS vibe, alien planet, vast, flying drones, background music + sound effects. Without any external assets at all. With proper game like graphics settings panel 😍 And i'm really impressed, because all this is generated with my mixed quant for Qwen3.8-27B, which gives me more context + speed for a really good quality work. DSK harness at beginning and Pi agent towards the end cause i wanted to get this done fast. Full quant name: HF aj9o9/Qwen3.8-27B-GGUF file name: Qwen3.8-27B-gdn8-q6attn-iq3ffn.gguf Just like all of you, i also hate having to compact context, so higher context without ever needing to compromise quality is what i'm always after and this quant is created for that. Gives me 210K context at q8_0 kv + dflash2 Q4_K_M for speculative decoding. This result in 23GB vram usage.

I'm really happy with Qwen3.8-27B ! All frustrations of not able to use Ox-Alpha and running out of credits for Glm-5.3, Grok turned me into Qwen-3.8-27b. Running locally on my 3090. This is the first fps game i'm creating with Qwen3.8-27b. I was after NMS vibe, alien planet, vast, flying drones, background music + sound effects. Without any external assets at all. With proper game like graphics settings panel 😍 And i'm really impressed, because all this is generated with my mixed quant for Qwen3.8-27B, which gives me more context + speed for a really good quality work. DSK harness at beginning and Pi agent towards the end cause i wanted to get this done fast. Full quant name: HF aj9o9/Qwen3.8-27B-GGUF file name: Qwen3.8-27B-gdn8-q6attn-iq3ffn.gguf Just like all of you, i also hate having to compact context, so higher context without ever needing to compromise quality is what i'm always after and this quant is created for that. Gives me 210K context at q8_0 kv + dflash2 Q4_K_M for speculative decoding. This result in 23GB vram usage.

AJ

12,696 görüntüleme • 1 ay önce

This happens all the time! This is why I feel like I'm not worth myself! This is why I feel like I cant grow! This is twitch actively nerfing my account every time I go live! The audience didn't leave they were still in the chat! When I asked twitch for help, I got an automated incomplete response that said "everything looks" . I am so tired of fighting this and have pleaded with twitch, Daniel Clancy and have told the VP FACE TO FACE about this issue and still nothing has been done! Why? I'm small potatoes no one cares and no one will even put an iota of time into solving this issue! This has been a constant every day for over 2 years and I cannot express how defeated I absolutely feel every time I press start to immediately have my audience ignored by twitch. I know some of you are going to say its not all about the numbers and if you enjoy it, it shouldn't matter. Unfortunately the ultimate truth is we are heavily judged on our CPM by everyone, potential partners, sponsors, collaborative opportunities and even by twitch themselves. So yes, it does matter. It matters a lot and worst part is no one, will take this issue seriously! I have literally 120 days of information from twitches 9th longest subathon (unrecognized of course as again small potatoes) that would absolutely show this being a consistent issue. I don't know what to do, I am at wits end here and I every time I see this happen to me I just want to curl up into a ball and let the world wash over me. I need help, i need someone to look at this, because I have dedicated my life to entertaining individuals and I do this full time, I dedicate 8 hours a day every day for the past 2 years to this company and the only thing I get is ignored. I know this is a lot and it looks like just some random person complaining, but I need some way to get this in the right hands and not the automated jarble that I've been getting. Its even been acknowledged in the past by a tech for twitch that something might be up. Again, I know I'm not big but were all working under the same logo and should all be taken care of equally! Twitch

Frosted Fricks 💣🐦‍⬛

21,589 görüntüleme • 2 yıl önce

watch bonsai 3.9gb model crush tool calls. this is running in hermes agent locally, and it doesn't just fire blind, it reasons first before every single action, thinks through what it actually needs, then makes the call. zero failures the whole run, not one dropped, not one malformed, across a full multi step agent loop. for anyone new here, bonsai is a 1 bit quant of qwen 3.6 27b dense, the exact model i spent all month calling the king of the 3090. PrismML took the king, crushed it down to a single sign bit per weight, 3.9gb, small enough to sit next to a gemma 12b, and it still holds an agent loop and reasons before it acts. sub 4 bit is where models are supposed to turn to mush. someone even clocked it at tool-calling 0 out of 100. not in my runs. this is the warm-up. the real test is the build gauntlet my q4 27b nailed, elements, particle swarm, mandelbrot, and openclaw runs the same one right after. we find out if 1bit actually keeps up.

watch bonsai 3.9gb model crush tool calls. this is running in hermes agent locally, and it doesn't just fire blind, it reasons first before every single action, thinks through what it actually needs, then makes the call. zero failures the whole run, not one dropped, not one malformed, across a full multi step agent loop. for anyone new here, bonsai is a 1 bit quant of qwen 3.6 27b dense, the exact model i spent all month calling the king of the 3090. PrismML took the king, crushed it down to a single sign bit per weight, 3.9gb, small enough to sit next to a gemma 12b, and it still holds an agent loop and reasons before it acts. sub 4 bit is where models are supposed to turn to mush. someone even clocked it at tool-calling 0 out of 100. not in my runs. this is the warm-up. the real test is the build gauntlet my q4 27b nailed, elements, particle swarm, mandelbrot, and openclaw runs the same one right after. we find out if 1bit actually keeps up.

Sudo su

55,466 görüntüleme • 2 ay önce

Seedance 2.0 vs Gemini Omni, tested under the same conditions 👀 I was working on a new storyboard. After generating it with Seedance 2.0, I gave the exact same prompt + storyboard reference + character reference to Gemini Omni as well. Result: Gemini Omni surprised me with style quality and got closer than I expected in prompt adherence. But Seedance still feels ahead for storyboard execution, motion energy, camera language and environmental interaction. Gemini looks good. Seedance feels directed. You can see them in the video and the storyboard sheet is shown at the end. Which one do you think is more useful for this kind of work?

Seedance 2.0 vs Gemini Omni, tested under the same conditions 👀 I was working on a new storyboard. After generating it with Seedance 2.0, I gave the exact same prompt + storyboard reference + character reference to Gemini Omni as well. Result: Gemini Omni surprised me with style quality and got closer than I expected in prompt adherence. But Seedance still feels ahead for storyboard execution, motion energy, camera language and environmental interaction. Gemini looks good. Seedance feels directed. You can see them in the video and the storyboard sheet is shown at the end. Which one do you think is more useful for this kind of work?

Kōda

36,257 görüntüleme • 4 ay önce

MTPLX V2.10 is out! I put a lot of work into this one. Highlights: - Qwen 3.8 Next Support Added: Bare Speed (4 bit) and Optimised Speed (4 bit/8 bit) models added running at 50 - 80 TPS sustained at high contexts. (expect speed improvements in the coming days) - Qwen 3.8 Next mmap: The 32 GB n-gram table streams from SSD via mmap instead of RAM. Both models fit a 96 GB Mac with 0 speed impact. - Qwen 3.8 27B Decode & Prefill boost EVERYWHERE: +15% TPS at 3k context, +29% at 88k, +54% at 147k. Prefill +41%. - 48 GB Macs went from 3 to 4 TPS in swap @ 30k context to 33 TPS. - KV Cache has been solved!: 8 bit and 4 bit cache now barely impacts speed instead of a 50% decline. - Coding agents wall time optimisations: file task dropped from 150s to 44s, and mid-session first token went from ~2s to 0.11s. - CLI UX improvements + 34 bug fixes. My thoughts on Qwen Next? Incredible model. Enormously better at design, and creative tasks with scary good vision. It created what I think is the best flappy bird. maybe even better than the Fable version. It had various difficulties and skins, moving pillars, and upgrades. It also had the most memeified look. Incredibly impressed. However, it is slightly worse than 27B at having code just work straight away. But it's personality is also far better than 27B. Here is the Flappy bird it made below! Made with Qwen Next Optimized Speed one shot@ 57 TPS:

MTPLX V2.10 is out! I put a lot of work into this one. Highlights: - Qwen 3.8 Next Support Added: Bare Speed (4 bit) and Optimised Speed (4 bit/8 bit) models added running at 50 - 80 TPS sustained at high contexts. (expect speed improvements in the coming days) - Qwen 3.8 Next mmap: The 32 GB n-gram table streams from SSD via mmap instead of RAM. Both models fit a 96 GB Mac with 0 speed impact. - Qwen 3.8 27B Decode & Prefill boost EVERYWHERE: +15% TPS at 3k context, +29% at 88k, +54% at 147k. Prefill +41%. - 48 GB Macs went from 3 to 4 TPS in swap @ 30k context to 33 TPS. - KV Cache has been solved!: 8 bit and 4 bit cache now barely impacts speed instead of a 50% decline. - Coding agents wall time optimisations: file task dropped from 150s to 44s, and mid-session first token went from ~2s to 0.11s. - CLI UX improvements + 34 bug fixes. My thoughts on Qwen Next? Incredible model. Enormously better at design, and creative tasks with scary good vision. It created what I think is the best flappy bird. maybe even better than the Fable version. It had various difficulties and skins, moving pillars, and upgrades. It also had the most memeified look. Incredibly impressed. However, it is slightly worse than 27B at having code just work straight away. But it's personality is also far better than 27B. Here is the Flappy bird it made below! Made with Qwen Next Optimized Speed one shot@ 57 TPS:

Youssof Al Toukhi

21,126 görüntüleme • 1 ay önce