Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

2x RTX 3090s running 32 coding agents in parallel. Task: Build a trading bot (handle partial fills, rate limits, race conditions + explain reasoning). 🚀 qwen3.6-35b-a3b-autoround on 3090 35B • Total: 76,183 tokens in 64s • Aggregate: 3051.1 tok/s peak · 1183.8 tok/s sustained • Per-stream: 101.2 high /... show more

Tech2Wild

1,975 subscribers

15,456 views • 26 days ago •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

This is WILD ! 2 x 3090s 🚀 qwen3.6-35b-a3b-autoround on 3090 35B 32 parallel coding agents • Total: 25,943 tokens in 17s • Aggregate: 252275.4 tok/s peak · 1548.1 tok/s sustained • Per-stream: 20000.0 high / 1666.7 low / 9567.7 avg tok/s • TTFT 0.43s avg · E2E 17s avg

This is WILD ! 2 x 3090s 🚀 qwen3.6-35b-a3b-autoround on 3090 35B 32 parallel coding agents • Total: 25,943 tokens in 17s • Aggregate: 252275.4 tok/s peak · 1548.1 tok/s sustained • Per-stream: 20000.0 high / 1666.7 low / 9567.7 avg tok/s • TTFT 0.43s avg · E2E 17s avg

Tech2Wild

14,587 views • 26 days ago

This is the FASTER MODEL I've EVER Ran LOCALLY 🚀 laguna on Laguna S 2.1 · 3090 Quad INT4 • 1 parallel coding agents • Total: 3,405 tokens in 14s • Aggregate: 256.0 tok/s peak · 249.9 tok/s sustained • Per-stream: 256.0 high / 256.0 low / 256.0 avg tok/s • TTFT 0.12s avg

This is the FASTER MODEL I've EVER Ran LOCALLY 🚀 laguna on Laguna S 2.1 · 3090 Quad INT4 • 1 parallel coding agents • Total: 3,405 tokens in 14s • Aggregate: 256.0 tok/s peak · 249.9 tok/s sustained • Per-stream: 256.0 high / 256.0 low / 256.0 avg tok/s • TTFT 0.12s avg

Tech2Wild

16,404 views • 11 days ago

Qwen 3.6. 35B. Running locally. 0.25s TTFT. 70 tok/s.

Qwen 3.6. 35B. Running locally. 0.25s TTFT. 70 tok/s.

Osaurus

101,562 views • 3 months ago

有人在 Qwen3.6 35B-A3B 发布当天，就在 4 款 GPU 上进行了基准测试。RTX 3090 - 49.78 tok/s RTX 4090 - 118.93 tok/s RTX 5090 - 160.37 tok/s DGX Spark - 59.98 tok/s 一个 35B 的模型能在消费级硬件上跑出这样的速度，对我来说依然觉得不可思议。

有人在 Qwen3.6 35B-A3B 发布当天，就在 4 款 GPU 上进行了基准测试。RTX 3090 - 49.78 tok/s RTX 4090 - 118.93 tok/s RTX 5090 - 160.37 tok/s DGX Spark - 59.98 tok/s 一个 35B 的模型能在消费级硬件上跑出这样的速度，对我来说依然觉得不可思议。

Bill The Investor

34,684 views • 3 months ago

How Fast is Gemma 4 on a MacBook Pro M4? Benchmarking Google's new MoE (26B-A4B) > Model size: 26.1 GiB > Load time: ~4.2s Comparing single request VS > concurrent requests performance > 32k total context, 4 parallel slots single request behavior > TTFT: 5.68s > prompt: 3,701 tokens @ 652 tok/s > decode: 40.08 tok/s sequential (1 request at a time): > avg duration: 20.5s > p99: 22.1s > throughput: 40.11 tok/s > clean finishes: 100% concurrent (4 parallel requests): > aggregate throughput: 47.25 tok/s > total system throughput: 262.27 tok/s > avg duration: 65.1s > p95 latency: 68.8s > req/sec: 0.058 Head-to-Head: Sequential vs Concurrent throughput: > 40.11 tok/s → 47.25 tok/s (+17.8%) > small gain despite 4x parallelism latency per request: > 20.5s → 65.1s (~3.2x slower) > you pay heavily for concurrency system throughput (true utilization): > ~40 tok/s → 262 tok/s (~6.5x total output) > this is where concurrency wins tokens per second (decode ceiling): > ~40 tok/s steady in both modes > hardware-bound, not scheduler-bound TTFT impact: > ~5.7s baseline → buried under queueing in concurrent > “headers waittime” becomes the bottleneck What this actually means? - You don’t get linear scaling from parallel slots - You trade latency for total output - Mac Unified Memory setup is clearly saturating - Bandwidth + Scheduling overhead show up immediately This is exactly why GPUs dominate here Concurrency without killing latency

How Fast is Gemma 4 on a MacBook Pro M4? Benchmarking Google's new MoE (26B-A4B) > Model size: 26.1 GiB > Load time: ~4.2s Comparing single request VS > concurrent requests performance > 32k total context, 4 parallel slots single request behavior > TTFT: 5.68s > prompt: 3,701 tokens @ 652 tok/s > decode: 40.08 tok/s sequential (1 request at a time): > avg duration: 20.5s > p99: 22.1s > throughput: 40.11 tok/s > clean finishes: 100% concurrent (4 parallel requests): > aggregate throughput: 47.25 tok/s > total system throughput: 262.27 tok/s > avg duration: 65.1s > p95 latency: 68.8s > req/sec: 0.058 Head-to-Head: Sequential vs Concurrent throughput: > 40.11 tok/s → 47.25 tok/s (+17.8%) > small gain despite 4x parallelism latency per request: > 20.5s → 65.1s (~3.2x slower) > you pay heavily for concurrency system throughput (true utilization): > ~40 tok/s → 262 tok/s (~6.5x total output) > this is where concurrency wins tokens per second (decode ceiling): > ~40 tok/s steady in both modes > hardware-bound, not scheduler-bound TTFT impact: > ~5.7s baseline → buried under queueing in concurrent > “headers waittime” becomes the bottleneck What this actually means? - You don’t get linear scaling from parallel slots - You trade latency for total output - Mac Unified Memory setup is clearly saturating - Bandwidth + Scheduling overhead show up immediately This is exactly why GPUs dominate here Concurrency without killing latency

Ahmad

88,866 views • 4 months ago

Ollama me daba 21 tok/s con Qwen3.6 35B (12 GB VRAM). Mismo modelo, misma GPU → llama.cpp + -ncmoe 15 = 70 tok/s. No es magia. Es un flag que Ollama no expone. Comando exacto: llama-cli -m ~/models/Qwen3.6-35B-A3B-UD-IQ3_XXS.gguf -ngl 99 -ncmoe 15 -p "Hola" Demo real aquí 👇

Ollama me daba 21 tok/s con Qwen3.6 35B (12 GB VRAM). Mismo modelo, misma GPU → llama.cpp + -ncmoe 15 = 70 tok/s. No es magia. Es un flag que Ollama no expone. Comando exacto: llama-cli -m ~/models/Qwen3.6-35B-A3B-UD-IQ3_XXS.gguf -ngl 99 -ncmoe 15 -p "Hola" Demo real aquí 👇

OscarMartin

302,902 views • 2 months ago

Everyone's comparing the DGX Spark to a 5090 and calling it slow. I think that's the wrong comparison. I ran Qwen3.6 35B-A3B FP8 with the full 262K context window enabled (~96GB RAM) — something gaming GPUs can't really do. Results: 🟢No context: 51.3 tok/s, TTFT 110ms 🟣200K prefill: 34.6 tok/s, TTFT 85s (~2,341 tok/s prefill) Prefill is way faster than a Mac. And 35 tok/s deep into 200K context, on a model this strong, is genuinely usable. The Spark plays a different game.

Everyone's comparing the DGX Spark to a 5090 and calling it slow. I think that's the wrong comparison. I ran Qwen3.6 35B-A3B FP8 with the full 262K context window enabled (~96GB RAM) — something gaming GPUs can't really do. Results: 🟢No context: 51.3 tok/s, TTFT 110ms 🟣200K prefill: 34.6 tok/s, TTFT 85s (~2,341 tok/s prefill) Prefill is way faster than a Mac. And 35 tok/s deep into 200K context, on a model this strong, is genuinely usable. The Spark plays a different game.

stevibe

34,109 views • 2 months ago

Laguna S 2.1 running native NVFP4 Picked Blackwell hardware for this side-by-side: 🟢 DGX Spark | 19.44 tok/s | 162ms TTFT 🔵 RTX PRO 6000 | 108.54 tok/s | 266ms TTFT 🟣 4× RTX 5090 (TP4+EP4) | 145.81 tok/s | 521ms TTFT 117.6B MoE (8.5B active), quantized to FP4, on your own box.

Laguna S 2.1 running native NVFP4 Picked Blackwell hardware for this side-by-side: 🟢 DGX Spark | 19.44 tok/s | 162ms TTFT 🔵 RTX PRO 6000 | 108.54 tok/s | 266ms TTFT 🟣 4× RTX 5090 (TP4+EP4) | 145.81 tok/s | 521ms TTFT 117.6B MoE (8.5B active), quantized to FP4, on your own box.

stevibe

22,889 views • 10 days ago

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 views • 1 month ago

Qwen3.6 35B-A3B dropped yesterday, so I ran it on 4 GPUs to see how it performs: 🟣 RTX 3090 — 49.78 tok/s, TTFT 852ms 🟡 RTX 4090 — 118.93 tok/s, TTFT 686ms 🟢 RTX 5090 — 160.37 tok/s, TTFT 409ms 🔵 DGX Spark — 59.98 tok/s, TTFT 228ms I went with ollama as the backend because honestly, it's the easiest way for most people to get started. One command, model pulled, done. I used Q4_K_M (24GB) across all four cards. The reason is the 3090 and 4090 don't support NVFP4 (only the 5090 and DGX Spark could use it). Keeping the same quant everywhere felt like the fairest way to compare. And yes, you can absolutely squeeze more performance out of every card with vLLM, SGLang, or TensorRT-LLM. But that's not what this test is about. This is just the out-of-the-box experience for folks who own a GPU and want to try the new model tonight.

Qwen3.6 35B-A3B dropped yesterday, so I ran it on 4 GPUs to see how it performs: 🟣 RTX 3090 — 49.78 tok/s, TTFT 852ms 🟡 RTX 4090 — 118.93 tok/s, TTFT 686ms 🟢 RTX 5090 — 160.37 tok/s, TTFT 409ms 🔵 DGX Spark — 59.98 tok/s, TTFT 228ms I went with ollama as the backend because honestly, it's the easiest way for most people to get started. One command, model pulled, done. I used Q4_K_M (24GB) across all four cards. The reason is the 3090 and 4090 don't support NVFP4 (only the 5090 and DGX Spark could use it). Keeping the same quant everywhere felt like the fairest way to compare. And yes, you can absolutely squeeze more performance out of every card with vLLM, SGLang, or TensorRT-LLM. But that's not what this test is about. This is just the out-of-the-box experience for folks who own a GPU and want to try the new model tonight.

stevibe

401,145 views • 3 months ago

GLM-5.1-478B-NVFP4 Running on: - 4x RTX Pro 6000 - Sglang - 370,000 max tokens (1.75x full context) - p10 27.7 | p90 45.6 tok/s decode (gen) - 1340 tok/s prefill I could get 2x decode if I limit to 64k context (100 tok/s) In this video it operates Figma (:

GLM-5.1-478B-NVFP4 Running on: - 4x RTX Pro 6000 - Sglang - 370,000 max tokens (1.75x full context) - p10 27.7 | p90 45.6 tok/s decode (gen) - 1340 tok/s prefill I could get 2x decode if I limit to 64k context (100 tok/s) In this video it operates Figma (:

0xSero

74,772 views • 3 months ago

Curious about ollama kimi-k2.6:cloud speed? 3 test runs: > 77.9 tok/s, TTFT 979ms > 114.3 tok/s, TTFT 788ms > 86.3 tok/s, TTFT 1117ms For comparison, OpenRouter stats: > Parasail: 14 tok/s > Moonshot AI: 27 tok/s > NovitaAI: 27 tok/s > Cloudflare: 71 tok/s Obvious caveat: cloud speeds fluctuate with load. Just sharing numbers for the curious.

Curious about ollama kimi-k2.6:cloud speed? 3 test runs: > 77.9 tok/s, TTFT 979ms > 114.3 tok/s, TTFT 788ms > 86.3 tok/s, TTFT 1117ms For comparison, OpenRouter stats: > Parasail: 14 tok/s > Moonshot AI: 27 tok/s > NovitaAI: 27 tok/s > Cloudflare: 71 tok/s Obvious caveat: cloud speeds fluctuate with load. Just sharing numbers for the curious.

stevibe

35,129 views • 3 months ago

The RTX 3090 is a 5-year-old GPU and it still runs a 27B model at 20 tok/s I tested Qwen3.5:27b across 3 generations of NVIDIA: 5090 → ~60 tok/s 4090 → ~40 tok/s 3090 → ~20 tok/s Perfectly linear scaling. Double the generation, double the speed.

The RTX 3090 is a 5-year-old GPU and it still runs a 27B model at 20 tok/s I tested Qwen3.5:27b across 3 generations of NVIDIA: 5090 → ~60 tok/s 4090 → ~40 tok/s 3090 → ~20 tok/s Perfectly linear scaling. Double the generation, double the speed.

stevibe

142,493 views • 4 months ago

DFlash v0.1.4 : custom Metal verify kernels for quantized Qwen3 hybrid models, plus significant peak memory reduction at long context. M5 Max 40-core GPU, 64GB, stock mlx_lm baseline: Qwen3.6-35B-A3B-4bit: ► @ 1024 · 138.3 → 300.3 tok/s (2.20x) ► @ 2048 · 135.6 → 246.4 tok/s (1.81x) ► @ 4096 · 134.5 → 208.4 tok/s (1.56x) ► @ 8192 · 133.2 → 177.4 tok/s (1.33x) Qwen3.5-27B-4bit: ► @ 1024 · 33.5 → 79.0 tok/s (2.37x) ► @ 2048 · 33.1 → 70.2 tok/s (2.12x) ► @ 4096 · 31.5 → 55.7 tok/s (1.77x) ► @ 8192 · 33.9 → 45.3 tok/s (1.34x) Working on making this usable for agentic workloads goal is to never drop below baseline at any context depth. LLM decode is memory-bandwidth bound. M5 Max runs at 614 GB/s, that's 1.5x more than M1-M4 Max (400-410 GB/s). Results will vary on lower bandwidth chips.

DFlash v0.1.4 : custom Metal verify kernels for quantized Qwen3 hybrid models, plus significant peak memory reduction at long context. M5 Max 40-core GPU, 64GB, stock mlx_lm baseline: Qwen3.6-35B-A3B-4bit: ► @ 1024 · 138.3 → 300.3 tok/s (2.20x) ► @ 2048 · 135.6 → 246.4 tok/s (1.81x) ► @ 4096 · 134.5 → 208.4 tok/s (1.56x) ► @ 8192 · 133.2 → 177.4 tok/s (1.33x) Qwen3.5-27B-4bit: ► @ 1024 · 33.5 → 79.0 tok/s (2.37x) ► @ 2048 · 33.1 → 70.2 tok/s (2.12x) ► @ 4096 · 31.5 → 55.7 tok/s (1.77x) ► @ 8192 · 33.9 → 45.3 tok/s (1.34x) Working on making this usable for agentic workloads goal is to never drop below baseline at any context depth. LLM decode is memory-bandwidth bound. M5 Max runs at 614 GB/s, that's 1.5x more than M1-M4 Max (400-410 GB/s). Results will vary on lower bandwidth chips.

bstn 👁️

23,120 views • 3 months ago

16 people chatting with Qwen3.6-35B at once ONE DGX Spark. This is a real capture, not a mockup: every token you see replays at its actual measured arrival time. Peaks ~440 tok/s total. A single user gets 105 tok/s. NVFP4 + MTP-3 on vLLM.

16 people chatting with Qwen3.6-35B at once ONE DGX Spark. This is a real capture, not a mockup: every token you see replays at its actual measured arrival time. Peaks ~440 tok/s total. A single user gets 105 tok/s. NVFP4 + MTP-3 on vLLM.

Wësche

26,094 views • 23 days ago

Qwen3.5-35B with only 3B active parameters This MoE model runs FASTER than most 7B dense models. Tested on 3 generations of NVIDIA: - 5090: 137 tok/s - 4090: 112 tok/s - 3090: 78 tok/s The surprise? The 4090 <> 5090 gap is only 22%. With a 3B active MoE, even old GPUs fly.

Qwen3.5-35B with only 3B active parameters This MoE model runs FASTER than most 7B dense models. Tested on 3 generations of NVIDIA: - 5090: 137 tok/s - 4090: 112 tok/s - 3090: 78 tok/s The surprise? The 4090 <> 5090 gap is only 22%. With a 3B active MoE, even old GPUs fly.

stevibe

69,532 views • 4 months ago

"Why are you benchmarking DGX Spark? It's a training box." Yeah. Low bandwidth, but 128GB of unified memory is just sitting there. Plenty of room to optimize. DGX Spark + Qwen3.6 27B. Four backend/quant combos: 🔴 llama.cpp + UD_Q4_K_XL > 11.0 tok/s (baseline), TTFT 297ms 🟢 llama.cpp + DFlash > 20.4 tok/s (peaks at 97 tok/s), TTFT 320ms 🟡 vLLM FP8 + MTP > 13.1 tok/s, TTFT 540ms 🟣 vLLM NVFP4 + MTP > 24.2 tok/s, TTFT 376ms NVFP4+MTP is the winner for me, rock stable around 24 tok/s, no wild swings. DFlash is the wildcard: massive peaks, but fluctuates a lot. FP8+MTP barely beats baseline, and it's FP8. Love my Spark.

"Why are you benchmarking DGX Spark? It's a training box." Yeah. Low bandwidth, but 128GB of unified memory is just sitting there. Plenty of room to optimize. DGX Spark + Qwen3.6 27B. Four backend/quant combos: 🔴 llama.cpp + UD_Q4_K_XL > 11.0 tok/s (baseline), TTFT 297ms 🟢 llama.cpp + DFlash > 20.4 tok/s (peaks at 97 tok/s), TTFT 320ms 🟡 vLLM FP8 + MTP > 13.1 tok/s, TTFT 540ms 🟣 vLLM NVFP4 + MTP > 24.2 tok/s, TTFT 376ms NVFP4+MTP is the winner for me, rock stable around 24 tok/s, no wild swings. DFlash is the wildcard: massive peaks, but fluctuates a lot. FP8+MTP barely beats baseline, and it's FP8. Love my Spark.

stevibe

40,118 views • 2 months ago

Ran Google’s cookbook with 10 agents on my tiny GB10 GPU. 436 tok/s / 43.6 per agent Qwen3.6-35B + Dflash + DDTree on vLLM GB10 @ 74W The future isn't 10,000 GPUs in a nuclear-powered data center. It’s 10 agents on your desk solving your problems while you make your coffee.

Ran Google’s cookbook with 10 agents on my tiny GB10 GPU. 436 tok/s / 43.6 per agent Qwen3.6-35B + Dflash + DDTree on vLLM GB10 @ 74W The future isn't 10,000 GPUs in a nuclear-powered data center. It’s 10 agents on your desk solving your problems while you make your coffee.

Mitko Vasilev

146,363 views • 3 months ago

Apple Silicon + Gemma 4 fans: this is for you. Pico AI Server now supports continuous batching with MLX-Swift. 43 tok/s on 1 stream. 26 tok/s per stream on 2 concurrent streams. That’s 52 tok/s total. a 21% throughput gain on a six-year-old MacBook Pro M1 Max!

Apple Silicon + Gemma 4 fans: this is for you. Pico AI Server now supports continuous batching with MLX-Swift. 43 tok/s on 1 stream. 26 tok/s per stream on 2 concurrent streams. That’s 52 tok/s total. a 21% throughput gain on a six-year-old MacBook Pro M1 Max!

Ronald Mannak

50,180 views • 3 months ago

Qwen3.5-35B-A3B running locally on an M4 chip at 49.5 tokens per second. A 35B model. On a laptop. In real time. LOCAL AI IS GETTING SCARY FAST.

Qwen3.5-35B-A3B running locally on an M4 chip at 49.5 tokens per second. A 35B model. On a laptop. In real time. LOCAL AI IS GETTING SCARY FAST.

0xMarioNawfal

478,178 views • 5 months ago