OmedTheVibeCoder's banner

OmedTheVibeCoder

@OmedVibeCodes • 1,441 subscribers

🇩🇪 Software Developer | Extreme AI model testing, honest results, no hype | Building something that could change game development 🎮

Shorts

GPT 5.6 Luna Max vs 5.6 Sol Medium… Yeah, it’s not even a question 😭 Sol on MEDIUM produced a much better result than Luna on MAX—and needed only around HALF the time. Same harness. Better generation. Faster result. Luna may be cheaper, but if you have the budget, I would absolutely use Sol. Saving money is not worth it when the weaker output creates hours of extra work later.

GPT 5.6 Luna Max vs 5.6 Sol Medium… Yeah, it’s not even a question 😭 Sol on MEDIUM produced a much better result than Luna on MAX—and needed only around HALF the time. Same harness. Better generation. Faster result. Luna may be cheaper, but if you have the budget, I would absolutely use Sol. Saving money is not worth it when the weaker output creates hours of extra work later.

100,364 görüntüleme

DeepSeek V4 Flash being ranked almost equal to Opus 4.8 is actually insane to me. Against Kimi K3, Opus 5, GPT-5.6 Sol, and Qwen 3.8 Max Preview, you give up a LOT by choosing DeepSeek. This is not a 2-point difference in practice. I’ll post the Opus 4.8 comparison next because you need to see this.

DeepSeek V4 Flash being ranked almost equal to Opus 4.8 is actually insane to me. Against Kimi K3, Opus 5, GPT-5.6 Sol, and Qwen 3.8 Max Preview, you give up a LOT by choosing DeepSeek. This is not a 2-point difference in practice. I’ll post the Opus 4.8 comparison next because you need to see this.

26,990 görüntüleme

Okay, I take back some of the DeepSeek V4 Flash hype 😭 My first generation was surprisingly good, but repeated tests through Pi produced some VERY questionable outputs. I’m even using the OpenCode Go plan, and it still didn’t impress me consistently. For me, this ended up being a flop.

Okay, I take back some of the DeepSeek V4 Flash hype 😭 My first generation was surprisingly good, but repeated tests through Pi produced some VERY questionable outputs. I’m even using the OpenCode Go plan, and it still didn’t impress me consistently. For me, this ended up being a flop.

21,015 görüntüleme

Videos

Anya Rossi

sweetdream.ai

SweetDream.ai•Sponsored•Livecam

Watch Anya Live

Anya is streaming live right now! Join her private show and enjoy exclusive content.

Exclusive private shows

1.2k viewers online

Private Show

Join now for exclusive access

Free preview available • Premium content

Just tested Qwen3.8-Max-Preview on my benchmark. It got absolutely cooked. Not Fable 5 level. Not even remotely close. It doesn't come close to GPT-5.6 or Opus 4.8 Max either. Honestly, marketing it at that level feels incredibly misleading. I wasted my time and money testing this. Qwen, refund me.

Just tested Qwen3.8-Max-Preview on my benchmark. It got absolutely cooked. Not Fable 5 level. Not even remotely close. It doesn't come close to GPT-5.6 or Opus 4.8 Max either. Honestly, marketing it at that level feels incredibly misleading. I wasted my time and money testing this. Qwen, refund me.

OmedTheVibeCoder

183,804 görüntüleme • 2 ay önce

Nah, this can’t be real. Same GPT-5.6 Sol. Same task. Same thinking level. I switched from regular Codex to the Claude Code harness, and the result instantly became way better. Claude Code is somehow making every model cook. GPT-5.6 SOL WAS ALREADY TOP TIER!! 😭

Nah, this can’t be real. Same GPT-5.6 Sol. Same task. Same thinking level. I switched from regular Codex to the Claude Code harness, and the result instantly became way better. Claude Code is somehow making every model cook. GPT-5.6 SOL WAS ALREADY TOP TIER!! 😭

OmedTheVibeCoder

127,040 görüntüleme • 2 ay önce

Muse Spark 1.3 is not that good, and Gemini Flash 3.8 is just pure benchmark-maxxing. Gemini even choked on the first run and needed a second attempt just to produce a working scene. By stripping colliders and enforcing rigorous backend-level logic in 3D space, this benchmark immediately separates true reasoning from trained test shortcuts.

Muse Spark 1.3 is not that good, and Gemini Flash 3.8 is just pure benchmark-maxxing. Gemini even choked on the first run and needed a second attempt just to produce a working scene. By stripping colliders and enforcing rigorous backend-level logic in 3D space, this benchmark immediately separates true reasoning from trained test shortcuts.

OmedTheVibeCoder

29,403 görüntüleme • 19 gün önce

This is probably my most important post about DeepSeek V4 Flash. Do NOT judge this model after testing only one harness. Pi, Claude Code, and OpenCode produced dramatically different results—and OpenCode completely changed my verdict. You really need to read the details below.

This is probably my most important post about DeepSeek V4 Flash. Do NOT judge this model after testing only one harness. Pi, Claude Code, and OpenCode produced dramatically different results—and OpenCode completely changed my verdict. You really need to read the details below.

OmedTheVibeCoder

68,681 görüntüleme • 1 ay önce

This is why my benchmark is not benchmaxxable and is one of the best on twitter Put Gemini 3.8 High against Opus 5 xHigh and the hype does not hold up. It got mogged hard. No real leap here, just obvious benchmark-maxxing I have heavy constraints in this test, so it clearly shows, if a model will be able to still mantain good quality, while having to follow architectural patterns

This is why my benchmark is not benchmaxxable and is one of the best on twitter Put Gemini 3.8 High against Opus 5 xHigh and the hype does not hold up. It got mogged hard. No real leap here, just obvious benchmark-maxxing I have heavy constraints in this test, so it clearly shows, if a model will be able to still mantain good quality, while having to follow architectural patterns

OmedTheVibeCoder

18,850 görüntüleme • 19 gün önce

Opus 5 High might be the most ridiculous model I’ve ever tested. It produced this result in around ONE HOUR. Fable 5 needed almost 3 hours. GPT-5.6 Sol worked for over an hour and still created almost nothing compared to this. Opus 5 didn’t just win. It completely OUTCLASSED them.

Opus 5 High might be the most ridiculous model I’ve ever tested. It produced this result in around ONE HOUR. Fable 5 needed almost 3 hours. GPT-5.6 Sol worked for over an hour and still created almost nothing compared to this. Opus 5 didn’t just win. It completely OUTCLASSED them.

OmedTheVibeCoder

39,729 görüntüleme • 1 ay önce

Okay… DeepSeek V4 Flash actually COOKED 😭 With OpenCode, I’d place it around GPT-5.6 Terra Max—and definitely above Luna Max in this test. Almost unlimited usage for around €9, by the way. This model is a MASSIVE success. Just don’t use the wrong harness.

Okay… DeepSeek V4 Flash actually COOKED 😭 With OpenCode, I’d place it around GPT-5.6 Terra Max—and definitely above Luna Max in this test. Almost unlimited usage for around €9, by the way. This model is a MASSIVE success. Just don’t use the wrong harness.

OmedTheVibeCoder

35,417 görüntüleme • 1 ay önce

$I think I accidentally created a completely new game genre. I’ve been working on this nonstop for months, and I’m finally ready to show just a tiny fraction of the gameplay. No traditional 3D models. Everything is generated in Three.js through pure math. This is completely insane guys. The actual main feature is still hidden—and it changes EVERYTHING. #threejs$

I think I accidentally created a completely new game genre. I’ve been working on this nonstop for months, and I’m finally ready to show just a tiny fraction of the gameplay. No traditional 3D models. Everything is generated in Three.js through pure math. This is completely insane guys. The actual main feature is still hidden—and it changes EVERYTHING. #threejs

OmedTheVibeCoder

37,476 görüntüleme • 1 ay önce

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.

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

36,074 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

DeepSeek V4 Flash completely failed this test. Compared to Opus 4.8, GPT-5.6 Sol, Kimi K3, and the other current top models, it’s honestly not good. The gap is massive. Against Luna, however, it performs similarly well. Both have different strengths and weaknesses, so I still need to test them more. The weird part: the Pi harness performed worse than OpenCode for DeepSeek. More tests are coming, guys. Don’t worry

DeepSeek V4 Flash completely failed this test. Compared to Opus 4.8, GPT-5.6 Sol, Kimi K3, and the other current top models, it’s honestly not good. The gap is massive. Against Luna, however, it performs similarly well. Both have different strengths and weaknesses, so I still need to test them more. The weird part: the Pi harness performed worse than OpenCode for DeepSeek. More tests are coming, guys. Don’t worry

OmedTheVibeCoder

26,166 görüntüleme • 1 ay önce

GLM 5.3 Max vs Fable 5 vs Qwen 3.8 Max vs Grok 4.6 GLM 5.3 is a massive jump over 5.2. Still not beating Fable 5 for me, but it’s way closer now. And vs Grok 4.6? I’d take GLM 5.3 pretty easily on this test.

GLM 5.3 Max vs Fable 5 vs Qwen 3.8 Max vs Grok 4.6 GLM 5.3 is a massive jump over 5.2. Still not beating Fable 5 for me, but it’s way closer now. And vs Grok 4.6? I’d take GLM 5.3 pretty easily on this test.

OmedTheVibeCoder

16,648 görüntüleme • 1 ay önce

You need to read this. Again. Qwen 3.8 Max Preview still performs badly in OpenCode and Qwen Code. Switch to the Claude Code harness, and it suddenly becomes dramatically better. This is reproducible. The harness difference is real.

You need to read this. Again. Qwen 3.8 Max Preview still performs badly in OpenCode and Qwen Code. Switch to the Claude Code harness, and it suddenly becomes dramatically better. This is reproducible. The harness difference is real.

OmedTheVibeCoder

18,655 görüntüleme • 2 ay önce

Opus 5 vs Grok 4.6 xHigh vs Gemini 3.6 Flash vs Kimi K3 Guys… the gap is actually ridiculous 😭 Opus 5 is in a completely different world here. Especially compared to Gemini 3.6 Flash — it’s not even remotely close.

Opus 5 vs Grok 4.6 xHigh vs Gemini 3.6 Flash vs Kimi K3 Guys… the gap is actually ridiculous 😭 Opus 5 is in a completely different world here. Especially compared to Gemini 3.6 Flash — it’s not even remotely close.

OmedTheVibeCoder

12,823 görüntüleme • 1 ay önce

Daha fazla içerik yok.