Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

Qwen 3.8 Max vs Opus 5, side-by-side. 🌀 The Inception top, spin, wobble, fall 💡 A hand-crank generator lighting a bulb Qwen 3.8 Max vs Claude Opus 5. Both rendered. One costs 8x more. Worth it? Watch and decide.

stevibe's profile picture

stevibe

27,597 subscribers

49,166 Aufrufe • vor 7 Tagen •via X (Twitter)

KunstWissenschaft & TechnologieGaming
LIVE

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

Claude Opus 4.7 vs 4.8 side-by-side canvas test
0:26

Claude Opus 4.7 vs 4.8 side-by-side canvas test

stevibe

476,059 Aufrufe • vor 2 Monaten

Qwen 3.8 new checkpoint "Kinsley" vs Claude Fable 5 High 🔥🔥 i didnt expected this from Qwen > photorealistic 3D cantilever house in Three.js this is just a sample one - i will share more.....
0:52

Qwen 3.8 new checkpoint "Kinsley" vs Claude Fable 5 High 🔥🔥 i didnt expected this from Qwen > photorealistic 3D cantilever house in Three.js this is just a sample one - i will share more.....

Harshith

25,773 Aufrufe • vor 11 Tagen

Qwen 3.8 Preview is quietly catching Kimi K3 on frontend + game design.And this is just the preview. Full 3.8 is going to be a beast. We tested Qwen 3.8 Preview against Kimi K3 on 3D modeling and animation. Ran both through the same 3 prompts. Both on Max effort. Qwen 3.8 Preview was dramatically faster across the board, 3-4x quicker on every test. Kimi K3 needed 2 hours on the final scene; Qwen wrapped it in 31 minutes. Test 1 (Engine Sim) Qwen 3.8 Preview: 25min Kimi K3: 1hr 30min Test 2 (Voxel Dino Printer) Qwen 3.8 Preview: 8min Kimi K3: 45min Test 3 (Garden Scene) Qwen 3.8 Preview: 31min Kimi K3: 2hrs
0:40

Qwen 3.8 Preview is quietly catching Kimi K3 on frontend + game design.And this is just the preview. Full 3.8 is going to be a beast. We tested Qwen 3.8 Preview against Kimi K3 on 3D modeling and animation. Ran both through the same 3 prompts. Both on Max effort. Qwen 3.8 Preview was dramatically faster across the board, 3-4x quicker on every test. Kimi K3 needed 2 hours on the final scene; Qwen wrapped it in 31 minutes. Test 1 (Engine Sim) Qwen 3.8 Preview: 25min Kimi K3: 1hr 30min Test 2 (Voxel Dino Printer) Qwen 3.8 Preview: 8min Kimi K3: 45min Test 3 (Garden Scene) Qwen 3.8 Preview: 31min Kimi K3: 2hrs

GMI Cloud

25,513 Aufrufe • vor 19 Tagen

i tried Opus 5.0 vs Fable 5 Max. Claude Opus 5.0 is quite the AGI beast in 3D, the texture and details are absolutely insane. Fable one shotted this though, could be improved. credit to Chetaslua for the Opus 5 code.
0:54

i tried Opus 5.0 vs Fable 5 Max. Claude Opus 5.0 is quite the AGI beast in 3D, the texture and details are absolutely insane. Fable one shotted this though, could be improved. credit to Chetaslua for the Opus 5 code.

sui

262,244 Aufrufe • vor 18 Tagen

here's a side by side comparison of a ray-tracing simulator by gpt-5 and claude 4.1 (opus) gpt-5 one-shotted this. claude genuinely blew me away with how badly it failed (and you all know how much i like claude)
1:42

here's a side by side comparison of a ray-tracing simulator by gpt-5 and claude 4.1 (opus) gpt-5 one-shotted this. claude genuinely blew me away with how badly it failed (and you all know how much i like claude)

Riley Coyote

42,687 Aufrufe • vor 1 Jahr

Qwen 3.8 Max beat Fable 5 at building 3D physics scenes for 7x cheaper! We gave two models the same task. Build three self-contained 3D scenes, each one HTML file with real physics that runs itself. Prompts: - A marble machine that lifts marbles up a wheel and drops them on a loop - A car factory assembly line - A sawmill cutting logs into planks Outputs: Qwen 3.8 Max: 46.6K tokens, $0.28 Fable 5: 38.7K tokens, $1.93 Qwen worked much harder on the details. Its textures and shadows look real. Fable 5 built a fast prototype that works and left it there. In the marble machine, Fable's marbles show up from nowhere at the top of the wheel. In Qwen's, you can see them go up. On the car line, Qwen's cars look much better and more real. Fable's sawmill is very strange. The saw looks wrong, and the log goes right through a beam before the saw cuts it. Atomic Chat will have day zero support to run Qwen 3.8 Max locally!
0:30

Sensitive content

Qwen 3.8 Max beat Fable 5 at building 3D physics scenes for 7x cheaper! We gave two models the same task. Build three self-contained 3D scenes, each one HTML file with real physics that runs itself. Prompts: - A marble machine that lifts marbles up a wheel and drops them on a loop - A car factory assembly line - A sawmill cutting logs into planks Outputs: Qwen 3.8 Max: 46.6K tokens, $0.28 Fable 5: 38.7K tokens, $1.93 Qwen worked much harder on the details. Its textures and shadows look real. Fable 5 built a fast prototype that works and left it there. In the marble machine, Fable's marbles show up from nowhere at the top of the wheel. In Qwen's, you can see them go up. On the car line, Qwen's cars look much better and more real. Fable's sawmill is very strange. The saw looks wrong, and the log goes right through a beam before the saw cuts it. Atomic Chat will have day zero support to run Qwen 3.8 Max locally!

atomic.chat

208,314 Aufrufe • vor 7 Tagen

qwen 3.8 max vs deepseek v4 flash 0731 vs kimi k3 vs gpt 5.6 sol – on rubik's cube and chess four frontier models built a rubik's cube stand and solved it, then built a chess board and played claude opus 5 on it the setup: Nous Research's hermes agent cli on OpenRouter tasks: 1. cube – build a 3d rubik's cube with a cli and a Three.js viewer, then solve an identical scrambled position on your own stand 2. chess – build a 3d chess stand, then play white against claude opus 5 as black, live, one move at a time. no engine, no solver, no opening book on either side. stockfish depth 14 grades every chess ply afterwards; neither player sees the score models: DeepSeek v4 flash 0731, OpenAI gpt-5.6 sol, Kimi.ai kimi k3, Qwen qwen 3.8 max gpt-5.6 sol and deepseek v4 flash solved their cubes – sol in 24 moves and seventeen seconds, deepseek in 32. qwen and kimi never got there, giving up at 96 and 207 moves then all four built chess stands and played white against claude opus 5 on them, and all four resigned: deepseek on move 13, sol on 19, kimi on 21, qwen holding out longest at 29 - build time, both stands #1 gpt-5.6 sol – 16m 43s #2 deepseek v4 flash – 97m 39s #3 kimi k3 – 166m 09s #4 qwen 3.8 max – 215m 08s - build attempts before a working stand #1 gpt-5.6 sol – 3 #2 qwen 3.8 max – 4 #3 kimi k3 – 4 #4 deepseek v4 flash – 5 - total tokens #1 gpt-5.6 sol – 6,713,754 #2 qwen 3.8 max – 17,272,507 #3 kimi k3 – 22,427,504 #4 deepseek v4 flash – 27,417,442 - total price #1 deepseek v4 flash – $0.557 #2 gpt-5.6 sol – $6.319 #3 qwen 3.8 max – $10.270 #4 kimi k3 – $16.667 observations: • deepseek v4 flash is the cheapest model here by a margin nobody else is near, and it got there while being the least efficient of the four. it burned 27.4m tokens – more than anyone, 5m more than kimi – and still finished both benchmarks for $0.557. that is $0.02 per million tokens against kimi's $0.74. it also needed the most passes to produce working stands, five, and that did not matter: all five deepseek passes together cost a thirtieth of kimi's two • so what deepseek cannot do is get it right the first time. what it can do is get it right the fifth time, for half a dollar. that is a different thing to be buying – not a good first draft, but the option to keep asking • gpt-5.6 sol is the opposite profile and the strongest of the four on pure efficiency. 16m 43s to build both stands, 6.7m tokens, three passes – under 40% of the next lowest token count and a quarter of deepseek's, on an eighth of qwen's clock. it also solved the cube fastest of anyone, 24 moves in seventeen seconds. sol is what you reach for when you want the answer now and can absorb $0.94 per million • sol's weakness is in what it does not check. its chess viewer deleted the capturing piece instead of the captured one, so pieces disappeared off the board mid-game – a defect the fifty-cent deepseek stand did not have. fast and terse turns out to be the same dial as fast and unverified • qwen 3.8 max is not the cheap open-weights option it gets treated as. $10.270 across the two benchmarks, second most expensive of the four, 18x deepseek, and by a distance the slowest – 215 minutes of build time, nearly thirteen times sol's. what the money buys is judgment: it played eighteen moves without a single error worth a hundredth of a pawn, then made exactly one bad move in the whole game, and averaged 44.6 centipawns lost across the longest game any of the four managed. it also could not solve a rubik's cube in 96 tries • kimi k3 is the one line with no reading that flatters it. most expensive at $16.667, last on the cube at 207 moves, last at chess at 478 centipawns lost per move. it is also the model that verified hardest – on the cube it wrote its own integrity check instead of trusting its output. that makes the result worse rather than better: the checking was real, and the reasoning underneath it still was not follow thehype. for 24/7 ai news, analysis and breakdowns
1:03

qwen 3.8 max vs deepseek v4 flash 0731 vs kimi k3 vs gpt 5.6 sol – on rubik's cube and chess four frontier models built a rubik's cube stand and solved it, then built a chess board and played claude opus 5 on it the setup: Nous Research's hermes agent cli on OpenRouter tasks: 1. cube – build a 3d rubik's cube with a cli and a Three.js viewer, then solve an identical scrambled position on your own stand 2. chess – build a 3d chess stand, then play white against claude opus 5 as black, live, one move at a time. no engine, no solver, no opening book on either side. stockfish depth 14 grades every chess ply afterwards; neither player sees the score models: DeepSeek v4 flash 0731, OpenAI gpt-5.6 sol, Kimi.ai kimi k3, Qwen qwen 3.8 max gpt-5.6 sol and deepseek v4 flash solved their cubes – sol in 24 moves and seventeen seconds, deepseek in 32. qwen and kimi never got there, giving up at 96 and 207 moves then all four built chess stands and played white against claude opus 5 on them, and all four resigned: deepseek on move 13, sol on 19, kimi on 21, qwen holding out longest at 29 - build time, both stands #1 gpt-5.6 sol – 16m 43s #2 deepseek v4 flash – 97m 39s #3 kimi k3 – 166m 09s #4 qwen 3.8 max – 215m 08s - build attempts before a working stand #1 gpt-5.6 sol – 3 #2 qwen 3.8 max – 4 #3 kimi k3 – 4 #4 deepseek v4 flash – 5 - total tokens #1 gpt-5.6 sol – 6,713,754 #2 qwen 3.8 max – 17,272,507 #3 kimi k3 – 22,427,504 #4 deepseek v4 flash – 27,417,442 - total price #1 deepseek v4 flash – $0.557 #2 gpt-5.6 sol – $6.319 #3 qwen 3.8 max – $10.270 #4 kimi k3 – $16.667 observations: • deepseek v4 flash is the cheapest model here by a margin nobody else is near, and it got there while being the least efficient of the four. it burned 27.4m tokens – more than anyone, 5m more than kimi – and still finished both benchmarks for $0.557. that is $0.02 per million tokens against kimi's $0.74. it also needed the most passes to produce working stands, five, and that did not matter: all five deepseek passes together cost a thirtieth of kimi's two • so what deepseek cannot do is get it right the first time. what it can do is get it right the fifth time, for half a dollar. that is a different thing to be buying – not a good first draft, but the option to keep asking • gpt-5.6 sol is the opposite profile and the strongest of the four on pure efficiency. 16m 43s to build both stands, 6.7m tokens, three passes – under 40% of the next lowest token count and a quarter of deepseek's, on an eighth of qwen's clock. it also solved the cube fastest of anyone, 24 moves in seventeen seconds. sol is what you reach for when you want the answer now and can absorb $0.94 per million • sol's weakness is in what it does not check. its chess viewer deleted the capturing piece instead of the captured one, so pieces disappeared off the board mid-game – a defect the fifty-cent deepseek stand did not have. fast and terse turns out to be the same dial as fast and unverified • qwen 3.8 max is not the cheap open-weights option it gets treated as. $10.270 across the two benchmarks, second most expensive of the four, 18x deepseek, and by a distance the slowest – 215 minutes of build time, nearly thirteen times sol's. what the money buys is judgment: it played eighteen moves without a single error worth a hundredth of a pawn, then made exactly one bad move in the whole game, and averaged 44.6 centipawns lost across the longest game any of the four managed. it also could not solve a rubik's cube in 96 tries • kimi k3 is the one line with no reading that flatters it. most expensive at $16.667, last on the cube at 207 moves, last at chess at 478 centipawns lost per move. it is also the model that verified hardest – on the cube it wrote its own integrity check instead of trusting its output. that makes the result worse rather than better: the checking was real, and the reasoning underneath it still was not follow thehype. for 24/7 ai news, analysis and breakdowns

thehype.

78,036 Aufrufe • vor 5 Tagen

3 ways to destroy a piece of paper. Qwen 3.5 35B A3B vs. Ornith 1.0 35B, side-by-side canvas test. (Why 3.5 not 3.6? Ornith is post-trained on Qwen 3.5 and Gemma 4, so this shows what the post-training adds.) Same 3 challenges: 🔪 Slice: three blade swipes, fruit-game style 📄 Shredder: desktop strip-cut 🗑️ Crumple: balled up and tossed Winner: not close. Ornith, decisively. The post-training quality is REAL.
0:20

3 ways to destroy a piece of paper. Qwen 3.5 35B A3B vs. Ornith 1.0 35B, side-by-side canvas test. (Why 3.5 not 3.6? Ornith is post-trained on Qwen 3.5 and Gemma 4, so this shows what the post-training adds.) Same 3 challenges: 🔪 Slice: three blade swipes, fruit-game style 📄 Shredder: desktop strip-cut 🗑️ Crumple: balled up and tossed Winner: not close. Ornith, decisively. The post-training quality is REAL.

stevibe

47,792 Aufrufe • vor 1 Monat

Guys, stop underestimating the harness. GPT-5.6 Sol through Claude Code vs Codex produced a GIGANTIC difference in my benchmark. But Opus 5 is the real innovation: High, xHigh, and Max don’t just feel like different thinking levels—they feel like COMPLETELY DIFFERENT MODELS. It’s basically like getting three or four Opus models in one.
0:43

Guys, stop underestimating the harness. GPT-5.6 Sol through Claude Code vs Codex produced a GIGANTIC difference in my benchmark. But Opus 5 is the real innovation: High, xHigh, and Max don’t just feel like different thinking levels—they feel like COMPLETELY DIFFERENT MODELS. It’s basically like getting three or four Opus models in one.

OmedTheVibeCoder

35,970 Aufrufe • vor 16 Tagen

Yesterday we found out Qwen3.5-27B (dense) beats both the 35B and 122B MoE variants at UI replication from screenshots. Today's question: what about Jackrong's Qwen3.5-27B-Claude-Opus-4.6-Reasoning-Distilled? A distilled 27B with Claude Opus 4.6 reasoning baked in. So I put all 3 in the ring: - Qwen3.5-27B - Qwen3.5-27B-Claude-Opus-Distilled - Claude Opus 4.6 Since the distilled model isn't multimodal, I gave all three the same detailed text prompts describing the UI components from yesterday's test.
0:31

Yesterday we found out Qwen3.5-27B (dense) beats both the 35B and 122B MoE variants at UI replication from screenshots. Today's question: what about Jackrong's Qwen3.5-27B-Claude-Opus-4.6-Reasoning-Distilled? A distilled 27B with Claude Opus 4.6 reasoning baked in. So I put all 3 in the ring: - Qwen3.5-27B - Qwen3.5-27B-Claude-Opus-Distilled - Claude Opus 4.6 Since the distilled model isn't multimodal, I gave all three the same detailed text prompts describing the UI components from yesterday's test.

stevibe

126,243 Aufrufe • vor 4 Monaten

"Its (Sonnet 5) performance is close to Opus 4.8, at lower prices." So I ran 4 canvas test through both. > Opus 4.8, 4/4 actually animating. > Sonnet 5, 2/4 came back as static images. And "lower price"? On the paper shredder task, Sonnet 5 spent $0.36 for a static image. Opus 4.8 spent $0.18 and it actually animated. The 4 tests: > Win 98 drag-to-BSOD > Self-typing keyboard + CRT > Letter burning > Paper shredder
0:34

"Its (Sonnet 5) performance is close to Opus 4.8, at lower prices." So I ran 4 canvas test through both. > Opus 4.8, 4/4 actually animating. > Sonnet 5, 2/4 came back as static images. And "lower price"? On the paper shredder task, Sonnet 5 spent $0.36 for a static image. Opus 4.8 spent $0.18 and it actually animated. The 4 tests: > Win 98 drag-to-BSOD > Self-typing keyboard + CRT > Letter burning > Paper shredder

stevibe

60,788 Aufrufe • vor 1 Monat

NVIDIA CEO JENSEN HUANG UNVEILED THE RTX 3090 AND THE USED VERSION ON EBAY NOW KILLS A $400/MONTH CLAUDE CODE AND OPENAI BILL the rtx 3090 has 24gb of vram, the same as a new rtx 4090 that costs $2,000. for local ai inference vram matters more than anything else and the 3090 sells for $650-800 used it runs qwen 3.6 27b comfortably, deepseek r1 14b at 60+ tokens per second and mistral 7b for fast automation. 16gb of qwen 3.6 fits with 8gb of headroom for long context alibaba's qwen 3.6 27b beats claude 4.5 opus on realworldqa vision 84.1 vs 77.0 and on ifbench instructions 76.5 vs 58.0. a free open weight model out-benchmarks anthropic's flagship a heavy ai user pays $200 for claude code max, $200 for chatgpt pro, $20 for cursor and $20 for gemini. that's $5,280 a year and a used 3090 pays itself off in 3.5 months install ollama, pull qwen 3.6 27b, point claude code at localhost. same interface, nothing leaves the machine, nothing costs per request bookmark this and read the article below
0:33

NVIDIA CEO JENSEN HUANG UNVEILED THE RTX 3090 AND THE USED VERSION ON EBAY NOW KILLS A $400/MONTH CLAUDE CODE AND OPENAI BILL the rtx 3090 has 24gb of vram, the same as a new rtx 4090 that costs $2,000. for local ai inference vram matters more than anything else and the 3090 sells for $650-800 used it runs qwen 3.6 27b comfortably, deepseek r1 14b at 60+ tokens per second and mistral 7b for fast automation. 16gb of qwen 3.6 fits with 8gb of headroom for long context alibaba's qwen 3.6 27b beats claude 4.5 opus on realworldqa vision 84.1 vs 77.0 and on ifbench instructions 76.5 vs 58.0. a free open weight model out-benchmarks anthropic's flagship a heavy ai user pays $200 for claude code max, $200 for chatgpt pro, $20 for cursor and $20 for gemini. that's $5,280 a year and a used 3090 pays itself off in 3.5 months install ollama, pull qwen 3.6 27b, point claude code at localhost. same interface, nothing leaves the machine, nothing costs per request bookmark this and read the article below

starmex

86,410 Aufrufe • vor 1 Monat

Crazy. Opus 5 outperformed Fable 5 on 3d modeling at 2/3 the price We tested Opus 5, Fable 5, and GPT 5.6 Sol on the same prompt: build an explorable Japanese suburban street in Three.js, full 3D, rendered like a hand-drawn anime background. All on Max effort. Opus 5 took the longest but delivered the most varied scenery: clean dark outlines, flat cel-shaded color bands, dense street details down to the vending machine signage. Fable paid more attention to details, and while GPT 5.6 Sol was 3-4x cheaper and the fastest, but the render shows it: sparse, washed out, barely styled. Opus 5: 1hr 21min · $26.7 Fable 5: 1hr 5min · $35 GPT 5.6 Sol: 41min · $7.1
0:34

Crazy. Opus 5 outperformed Fable 5 on 3d modeling at 2/3 the price We tested Opus 5, Fable 5, and GPT 5.6 Sol on the same prompt: build an explorable Japanese suburban street in Three.js, full 3D, rendered like a hand-drawn anime background. All on Max effort. Opus 5 took the longest but delivered the most varied scenery: clean dark outlines, flat cel-shaded color bands, dense street details down to the vending machine signage. Fable paid more attention to details, and while GPT 5.6 Sol was 3-4x cheaper and the fastest, but the render shows it: sparse, washed out, barely styled. Opus 5: 1hr 21min · $26.7 Fable 5: 1hr 5min · $35 GPT 5.6 Sol: 41min · $7.1

GMI Cloud

53,240 Aufrufe • vor 16 Tagen

This is F**king dangerous Someone figured out how to get dramatically sharper reasoning out of Opus 4.8 with one setup. Here is how: Ask Claude to write you an "operating manual" for how it should reason through your work Create a new Claude Project and paste that .md into the Project instructions Run Opus 4.8 on High or Max effort Now the model reasons against a real framework instead of improvising every response. Same model. Far better output. And Opus 4.8 already costs a fraction of Fable 5. Save and bookmark this. Follow CyrilXBT
0:41

This is F**king dangerous Someone figured out how to get dramatically sharper reasoning out of Opus 4.8 with one setup. Here is how: Ask Claude to write you an "operating manual" for how it should reason through your work Create a new Claude Project and paste that .md into the Project instructions Run Opus 4.8 on High or Max effort Now the model reasons against a real framework instead of improvising every response. Same model. Far better output. And Opus 4.8 already costs a fraction of Fable 5. Save and bookmark this. Follow CyrilXBT

CyrilXBT

198,085 Aufrufe • vor 1 Monat

CLAUDE SONNET 5 vs. CLAUDE OPUS 4.8 RESULTS Prompt: "build a single html landing page about claude sonnet 5, the new AI model that just dropped from Anthropic." Claude Sonnet 5: - Tokens Used: 20.9k input, 14.2k output - Total Cost: $3.36 - Time elapsed: 2m 11s Claude Opus 4.8 Ultracode: - Tokens Used: 96.3k input, 73.8k output - Total cost: $20.66 - Time Elapsed: 20m 15s I ran both of them in UltraCode mode via the CLI. I think in terms of quality of output, Opus won. In terms of model speed and cost, Sonnet 5 won. What are your thoughts? Comment down below who you think won!
0:25

CLAUDE SONNET 5 vs. CLAUDE OPUS 4.8 RESULTS Prompt: "build a single html landing page about claude sonnet 5, the new AI model that just dropped from Anthropic." Claude Sonnet 5: - Tokens Used: 20.9k input, 14.2k output - Total Cost: $3.36 - Time elapsed: 2m 11s Claude Opus 4.8 Ultracode: - Tokens Used: 96.3k input, 73.8k output - Total cost: $20.66 - Time Elapsed: 20m 15s I ran both of them in UltraCode mode via the CLI. I think in terms of quality of output, Opus won. In terms of model speed and cost, Sonnet 5 won. What are your thoughts? Comment down below who you think won!

Mark Santos

325,767 Aufrufe • vor 1 Monat

Claude Opus 4.8 (Max) is definitely capable, but this test honestly surprised me. I gave it the same macOS-style web OS challenge and it spent 1 hour and 56 minutes, burned through 66k+ tokens, and still produced a result that was worse than what Qwen 3.7-Max generated. The biggest issue isn't quality, it's efficiency. Opus 4.8 can absolutely cook when given enough time, but the token usage and runtime are becoming harder and harder to justify when cheaper models are delivering comparable or even better results. Full Breakdown: The frontier is getting very competitive. 👀
1:14

Claude Opus 4.8 (Max) is definitely capable, but this test honestly surprised me. I gave it the same macOS-style web OS challenge and it spent 1 hour and 56 minutes, burned through 66k+ tokens, and still produced a result that was worse than what Qwen 3.7-Max generated. The biggest issue isn't quality, it's efficiency. Opus 4.8 can absolutely cook when given enough time, but the token usage and runtime are becoming harder and harder to justify when cheaper models are delivering comparable or even better results. Full Breakdown: The frontier is getting very competitive. 👀

WorldofAI

123,550 Aufrufe • vor 2 Monaten

Alright, here's a side-by-side comparison of OpenAI's new Codex vs Codebuff Codex took 5 minutes and didn't implement the client side of my request 😢 Codebuff took 1 minute and got it right! 🔥
5:17

Alright, here's a side-by-side comparison of OpenAI's new Codex vs Codebuff Codex took 5 minutes and didn't implement the client side of my request 😢 Codebuff took 1 minute and got it right! 🔥

James Grugett

43,750 Aufrufe • vor 1 Jahr

GPT-5.4 is my new default model for OpenClaw. For the last 2 months, Opus was the only model I felt could really unlock OpenClaw for serious agent tasks. GPT-5.4 is the first one that got close enough in quality to change that. In practice, it feels ~60% faster (73 tok/s vs 46), costs about half as much, and gives ~5x longer context. That combination makes OpenClaw much more snappy as a default. I tested it side by side against Opus 4.6 on deep research + slide generation, and the outputs are inspectable too. Full video below:
7:21

GPT-5.4 is my new default model for OpenClaw. For the last 2 months, Opus was the only model I felt could really unlock OpenClaw for serious agent tasks. GPT-5.4 is the first one that got close enough in quality to change that. In practice, it feels ~60% faster (73 tok/s vs 46), costs about half as much, and gives ~5x longer context. That combination makes OpenClaw much more snappy as a default. I tested it side by side against Opus 4.6 on deep research + slide generation, and the outputs are inspectable too. Full video below:

Quinn Leng

131,551 Aufrufe • vor 5 Monaten

... QwQ-Max-Preview Qwen Chat: Blog: 🤔 Today we release "Thinking (QwQ)" in Qwen Chat, backed by our QwQ-Max-Preview, which is a reasoning model based on Qwen2.5-Max. This model is still for preview. It is highly capable of math understanding, coding, agent, etc. Compared with Qwen2.5-Max, it is much smarter and has much more creatitvity. Very soon, we are about to release the official version of QwQ-Max, and we will open-weight both QwQ-Max and Qwen2.5-Max under the license of Apache 2.0! Furthermore, we will also provide smaller variants, e.g., QwQ-32B, which can be deployed on local devices. Also, since a great number of users are expecting our APP, we are going to release an Android and iOS APP while we release our official QwQ-Max. Stay tuned for better things from Qwen!
1:20

... QwQ-Max-Preview Qwen Chat: Blog: 🤔 Today we release "Thinking (QwQ)" in Qwen Chat, backed by our QwQ-Max-Preview, which is a reasoning model based on Qwen2.5-Max. This model is still for preview. It is highly capable of math understanding, coding, agent, etc. Compared with Qwen2.5-Max, it is much smarter and has much more creatitvity. Very soon, we are about to release the official version of QwQ-Max, and we will open-weight both QwQ-Max and Qwen2.5-Max under the license of Apache 2.0! Furthermore, we will also provide smaller variants, e.g., QwQ-32B, which can be deployed on local devices. Also, since a great number of users are expecting our APP, we are going to release an Android and iOS APP while we release our official QwQ-Max. Stay tuned for better things from Qwen!

Qwen

430,736 Aufrufe • vor 1 Jahr

Parameter scaling just broke in front of me. I gave 90 math problems as images to 10 local vision models, 3 runs each, only consistent answers across all 3 runs counted. Two findings: > Gemma 4 was the most consistent family, 31B took the crown at 89.6%. > But Qwen 3.5 4B came 2 answers behind it. A 4B. Ranked 2nd out of 10. Vision-math isn't one skill, it's two: reading the image, then solving it. The real lesson for anyone running local: small ≠ weak. If you're building agentic workflows, finding the right model for each task matters more than reaching for the biggest one. In this test, the 4B ran far faster because of its size, scored higher, and left VRAM free for the rest of your stack. Full results: 🥇 Gemma 4 31B — 242/270 (89.6%) 🥈 Qwen 3.5 4B — 240/270 (88.9%) 🥉 Gemma 4 E4B — 222/270 (82.2%) 🥉 Qwen 3.6 27B — 222/270 (82.2%) 5. Gemma 4 26B A4B — 216/270 (80.0%) 6. Qwen 3.5 2B — 201/270 (74.4%) 7. Gemma 4 E2B — 192/270 (71.1%) 8. Qwen 3.6 35B A3B — 192/270 (71.1%) 9. Qwen 3.5 9B — 168/270 (62.2%) 10. Qwen 3.5 0.8B — 45/270 (16.7%) All GGUF + mmproj, Unsloth's Q6_K_XL quant.
0:44

Parameter scaling just broke in front of me. I gave 90 math problems as images to 10 local vision models, 3 runs each, only consistent answers across all 3 runs counted. Two findings: > Gemma 4 was the most consistent family, 31B took the crown at 89.6%. > But Qwen 3.5 4B came 2 answers behind it. A 4B. Ranked 2nd out of 10. Vision-math isn't one skill, it's two: reading the image, then solving it. The real lesson for anyone running local: small ≠ weak. If you're building agentic workflows, finding the right model for each task matters more than reaching for the biggest one. In this test, the 4B ran far faster because of its size, scored higher, and left VRAM free for the rest of your stack. Full results: 🥇 Gemma 4 31B — 242/270 (89.6%) 🥈 Qwen 3.5 4B — 240/270 (88.9%) 🥉 Gemma 4 E4B — 222/270 (82.2%) 🥉 Qwen 3.6 27B — 222/270 (82.2%) 5. Gemma 4 26B A4B — 216/270 (80.0%) 6. Qwen 3.5 2B — 201/270 (74.4%) 7. Gemma 4 E2B — 192/270 (71.1%) 8. Qwen 3.6 35B A3B — 192/270 (71.1%) 9. Qwen 3.5 9B — 168/270 (62.2%) 10. Qwen 3.5 0.8B — 45/270 (16.7%) All GGUF + mmproj, Unsloth's Q6_K_XL quant.

stevibe

22,882 Aufrufe • vor 2 Monaten