正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

Gemma 4 12B C++ Skate game through opencode. Done with a q8 quant.

BijanBowen

5,627 subscribers

11,034 次观看 • 5 个月前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

Qwen3.8 max NYC skate game in a single C++ file

Qwen3.8 max NYC skate game in a single C++ file

BijanBowen

28,168 次观看 • 3 个月前

Ever wonder what makes Gemma 4 12B so unique? In this 2 minute overview series, learn what an encoder-free model is and how Gemma 4 12B processes image and audio inputs differently from the rest of the Gemma family!

Ever wonder what makes Gemma 4 12B so unique? In this 2 minute overview series, learn what an encoder-free model is and how Gemma 4 12B processes image and audio inputs differently from the rest of the Gemma family!

Google Gemma

43,450 次观看 • 2 个月前

A model reading the paper that invented it. Gemma 4 12B. From a screenshot. Local.

A model reading the paper that invented it. Gemma 4 12B. From a screenshot. Local.

Osaurus

23,076 次观看 • 5 个月前

Unlock local, agentic workflows with Gemma 4 12B and Google AI Edge, directly on your laptop. Experience 100% on-device AI: • Generate code in AI Edge Gallery (new to Mac) • Dictate and edit text via AI Edge Eloquent (new to Mac) • Serve Gemma 4 12B locally with LiteRT-LM Dive in:

Unlock local, agentic workflows with Gemma 4 12B and Google AI Edge, directly on your laptop. Experience 100% on-device AI: • Generate code in AI Edge Gallery (new to Mac) • Dictate and edit text via AI Edge Eloquent (new to Mac) • Serve Gemma 4 12B locally with LiteRT-LM Dive in:

Google for Developers

141,165 次观看 • 5 个月前

Gemma4 12B with Unsloth's Quant on DGX Spark Quants: - UD_Q4_K_XL - UD_Q5_K_XL - UD_Q6_K_XL - UD_Q8_K_XL Summary: - Q4: 25.21 tok/s, TTFT 168ms - Q5: 21.7 tok/s, TTFT 182ms - Q6: 17.68 tok/s, TTFT 193.95ms - Q8: 15.22 tok/s, TTFT 221ms

Gemma4 12B with Unsloth's Quant on DGX Spark Quants: - UD_Q4_K_XL - UD_Q5_K_XL - UD_Q6_K_XL - UD_Q8_K_XL Summary: - Q4: 25.21 tok/s, TTFT 168ms - Q5: 21.7 tok/s, TTFT 182ms - Q6: 17.68 tok/s, TTFT 193.95ms - Q8: 15.22 tok/s, TTFT 221ms

stevibe

18,639 次观看 • 5 个月前

A long time coming but new mlx-lm is here with better batching support in the server and Gemma 4. pip install -U mlx-lm Here is a video where a single M3 Ultra serves 5 opencode sessions with Gemma 4 26B that process ~130k tokens in ~1.5 minutes.

A long time coming but new mlx-lm is here with better batching support in the server and Gemma 4. pip install -U mlx-lm Here is a video where a single M3 Ultra serves 5 opencode sessions with Gemma 4 26B that process ~130k tokens in ~1.5 minutes.

Angelos Katharopoulos

66,202 次观看 • 7 个月前

New Google Gemma 4 12B claims near-26B performance - we tested both! We ran both models locally on one RTX 4090 and gave each the same task: write a self-contained HTML5 canvas animation with real physics in one file without libraries. Three scenes - a Galton board, two blocks colliding off a wall, and a chaotic triple pendulum Outputs: Gemma 4 26B-A4B: 15 GB VRAM usage, 6.9k tokens, 138 tok/s Gemma 4 12B: 9 GB VRAM usage, 8.9k tokens, 80 tok/s Same Gemma 4 family, but the 26B-A4B won every scene and ran ~1.7x faster - on just 4B active params. The 12B stayed very close though, on almost half the VRAM - which makes it the ideal model for a 16 GB laptop

New Google Gemma 4 12B claims near-26B performance - we tested both! We ran both models locally on one RTX 4090 and gave each the same task: write a self-contained HTML5 canvas animation with real physics in one file without libraries. Three scenes - a Galton board, two blocks colliding off a wall, and a chaotic triple pendulum Outputs: Gemma 4 26B-A4B: 15 GB VRAM usage, 6.9k tokens, 138 tok/s Gemma 4 12B: 9 GB VRAM usage, 8.9k tokens, 80 tok/s Same Gemma 4 family, but the 26B-A4B won every scene and ran ~1.7x faster - on just 4B active params. The 12B stayed very close though, on almost half the VRAM - which makes it the ideal model for a 16 GB laptop

atomic.chat

152,748 次观看 • 5 个月前

And here are Qwen 3.8 27b Q8 websites definitely better than the NVFP4. Also getting really good game generation with Q8. Q8 is a win.

And here are Qwen 3.8 27b Q8 websites definitely better than the NVFP4. Also getting really good game generation with Q8. Q8 is a win.

Loktar 🇺🇸

11,512 次观看 • 2 个月前

$Gemma 4 12B QAT + MTP on 8GB VRAM. llama.cpp flags included. let's run it. 20+ tok/sec decode. 700+ tok/sec prefill. on a single RTX 4060. copy these exact flags: -m gemma-4-12B-it-qat-UD-Q4_K_XL.gguf \ --spec-type draft-mtp \ --spec-draft-n-max 4 \ --spec-draft-p-min 0.7 \ --spec-draft-model gemma-4-12B-it-qat-assistant-MTP-Q8_0.gguf \ -c 48000 -ngl 38 -v → -ngl 99 if you're on 12–24GB VRAM (RTX 3090, 4090, 4080, 3080, 4070 Ti, 3080 Ti) → -ngl 38 for 8GB setups with 48k context (RTX 4060, 3060, 2080, 2070, 3070) → drop it (or the context) lower if you're squeezing on 4–6GB (RTX 2060, 3050, 1660) my rig: RTX 4060 8GB · i7H · 16GB RAM MTP is giving me 25–40% decode throughput gains across Gemma 4 models. nearly zero VRAM cost for that bump. the draft assistant GGUF is only ~300–400MB depending on quant. one thing to know (architectural catch): unlike Qwen3.6 and Qwen3.5 models which bakes MTP heads straight into the base GGUF, Gemma 4 needs a separate draft assistant model downloaded alongside. not a big deal. just don't forget it or MTP won't run. draft assistant GGUF link → comments while you wait for anthropic mythos release, test this and drop your decode numbers below, curious how it scales across different setups.$

Gemma 4 12B QAT + MTP on 8GB VRAM. llama.cpp flags included. let's run it. 20+ tok/sec decode. 700+ tok/sec prefill. on a single RTX 4060. copy these exact flags: -m gemma-4-12B-it-qat-UD-Q4_K_XL.gguf \ --spec-type draft-mtp \ --spec-draft-n-max 4 \ --spec-draft-p-min 0.7 \ --spec-draft-model gemma-4-12B-it-qat-assistant-MTP-Q8_0.gguf \ -c 48000 -ngl 38 -v → -ngl 99 if you're on 12–24GB VRAM (RTX 3090, 4090, 4080, 3080, 4070 Ti, 3080 Ti) → -ngl 38 for 8GB setups with 48k context (RTX 4060, 3060, 2080, 2070, 3070) → drop it (or the context) lower if you're squeezing on 4–6GB (RTX 2060, 3050, 1660) my rig: RTX 4060 8GB · i7H · 16GB RAM MTP is giving me 25–40% decode throughput gains across Gemma 4 models. nearly zero VRAM cost for that bump. the draft assistant GGUF is only ~300–400MB depending on quant. one thing to know (architectural catch): unlike Qwen3.6 and Qwen3.5 models which bakes MTP heads straight into the base GGUF, Gemma 4 needs a separate draft assistant model downloaded alongside. not a big deal. just don't forget it or MTP won't run. draft assistant GGUF link → comments while you wait for anthropic mythos release, test this and drop your decode numbers below, curious how it scales across different setups.

Alok

29,006 次观看 • 5 个月前

ICYMI 👀: Google released a new Google AI Edge Foresight app for macOS, powered by Gemma 4 and EmbeddingGemma2 for recording voice notes. > The app can connect to your Google Drive in order to build a knowledge graph. > When transcription is active, it uses local models, Gemma 4 E4B or Gemma 4 12B, to transcribe voice notes and convert them into a new document.

ICYMI 👀: Google released a new Google AI Edge Foresight app for macOS, powered by Gemma 4 and EmbeddingGemma2 for recording voice notes. > The app can connect to your Google Drive in order to build a knowledge graph. > When transcription is active, it uses local models, Gemma 4 E4B or Gemma 4 12B, to transcribe voice notes and convert them into a new document.

🚨 AI News | TestingCatalog

37,600 次观看 • 1 个月前

Gemma 4 12B dropped today. Apache 2.0, multimodal: text, image, audio, and video. 256K context, built-in thinking, native tool calling. Running on Red Hat OpenShift AI with vLLM on Day 0:

Gemma 4 12B dropped today. Apache 2.0, multimodal: text, image, audio, and video. 256K context, built-in thinking, native tool calling. Running on Red Hat OpenShift AI with vLLM on Day 0:

Red Hat AI

15,995 次观看 • 5 个月前

I tested Gemma 4 12b on my Macbook Pro M5 Max with 128GB of unified memory. It runs VERY slow for a 12b parameter model. 44 tokens/second. The lava lamp that it produced was also terrible. I would stick to Qwen 3.6 27b or Qwen 3.6 35b. Google is still way behind, even with their local models.

I tested Gemma 4 12b on my Macbook Pro M5 Max with 128GB of unified memory. It runs VERY slow for a 12b parameter model. 44 tokens/second. The lava lamp that it produced was also terrible. I would stick to Qwen 3.6 27b or Qwen 3.6 35b. Google is still way behind, even with their local models.

BridgeMind

70,626 次观看 • 5 个月前

Running Gemma 4 12B locally at 28T/s. Facing issues with vision in LM Studio, but text performance is quite solid. Asked it to create a landing page for a tech event. Video is fast forwarded for the demo.

Running Gemma 4 12B locally at 28T/s. Facing issues with vision in LM Studio, but text performance is quite solid. Asked it to create a landing page for a tech event. Video is fast forwarded for the demo.

AshutoshShrivastava

15,697 次观看 • 5 个月前

2-bit Gemma 4 12B GGUF, only 4.66 GB on disk, managed to cite 15 sites from a single prompt. Try this locally on >6GB RAM via Unsloth Studio. GitHub:

2-bit Gemma 4 12B GGUF, only 4.66 GB on disk, managed to cite 15 sites from a single prompt. Try this locally on >6GB RAM via Unsloth Studio. GitHub:

Unsloth AI

145,438 次观看 • 5 个月前

Fable 5 UltraCode game full play through. All assets generated by Fable

Fable 5 UltraCode game full play through. All assets generated by Fable

BijanBowen

71,899 次观看 • 4 个月前

Trailer for 'Skate Mates', a multiplayer game where each player controls one skate, meaning you'll need teamwork to get through the tough obstacle courses

Trailer for 'Skate Mates', a multiplayer game where each player controls one skate, meaning you'll need teamwork to get through the tough obstacle courses

Dexerto

165,154 次观看 • 6 个月前

Fable 5 Wii game, had it port my castle time game to work on wii with wiimote. Still a WIP

Fable 5 Wii game, had it port my castle time game to work on wii with wiimote. Still a WIP

BijanBowen

20,294 次观看 • 4 个月前

Optional morning skate ahead of game 4. The Captain:

Optional morning skate ahead of game 4. The Captain:

Romi Bean

68,841 次观看 • 5 个月前

Meet Gemma Translator: a completely offline, DIY translation device powered by Gemma 4 and built with Antigravity. 🧵

Meet Gemma Translator: a completely offline, DIY translation device powered by Gemma 4 and built with Antigravity. 🧵

Google Antigravity

5,454,146 次观看 • 3 个月前