Qwen3 42B A3B 2507 Thinking Abliterated Uncensored TOTAL RECALL v2 Medium MASTER CODER — Hardware Requirements & GPU Compatibility
ChatCodeReasoning
Qwen3 42B A3B 2507 Thinking Abliterated Uncensored TOTAL RECALL v2 Medium MASTER CODER is a 42.4B-parameter open language model from DavidAU in the Qwen 3 family. It supports a context window of up to 262,144 tokens. At Q4_K_M it needs about 25.86 GB of VRAM — see which GPUs and Macs can run it below.
Runs well on a 24 GB GPU
Runs on 24 GB (Q3_K_M · ~142 tok/s · 32K)Macs from 24 GB (Q2_K · ~45 tok/s)
Apache 2.0
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Qwen3 42B A3B 2507 Thinking Abliterated Uncensored TOTAL RECALL v2 Medium MASTER CODER
One recommended way. Every other method, OS and quant is in the step-by-step guide.
Recommended
vllm serve DavidAU/Qwen3-42B-A3B-2507-Thinking-Abliterated-uncensored-TOTAL-RECALL-v2-Medium-MASTER-CODER --max-model-len 16384vLLM repo with its own quantisation
Download: Weights on Hugging Face
Other ways
each opens step-by-stepRun in cloud
Fits on A40 (48 GB) (21 GB headroom) · Q4_K_M
- Generation speed
- ~103 tok/s
- generation speed
- Cost per 1M output tokens
- $0.94
- per 1M output tokens
How Much VRAM Does Qwen3 42B A3B 2507 Thinking Abliterated Uncensored TOTAL RECALL v2 Medium MASTER CODER Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 18.4 GB | 20.6 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 64K context · ~151 tok/s | |
| 21.1 GB | 23.2 GB | 3-bit medium — 3-bit medium quantization | Fits: 32K context · ~142 tok/s | |
| 25.9 GB | 28 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Does not fit: needs 26.8 GB | |
| 30.6 GB | 32.7 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Does not fit: needs 31.6 GB | |
| 35.4 GB | 37.5 GB | 6-bit, very good — 6-bit quantization, very good quality | Does not fit: needs 36.4 GB | |
| 42.8 GB | 44.9 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Does not fit: needs 43.8 GB | |
| 85.2 GB | 87.3 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Does not fit: needs 86.2 GB |
Selected Q4_K_M
4-bit medium quantization26.8GBat 16K context
- Weights
- 25.4 GB
- KV cache
- 1.1 GB
- Overhead
- 0.3 GB
Doesn't fit RTX 4090 (reference) · needs 2.8 GB more, Q3_K_M fits
Which Hardware Can Run Qwen3 42B A3B 2507 Thinking Abliterated Uncensored TOTAL RECALL v2 Medium MASTER CODER?
Q4_K_M · 25.9 GBQwen3 42B A3B 2507 Thinking Abliterated Uncensored TOTAL RECALL v2 Medium MASTER CODER (Q4_K_M) requires 25.9 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 34+ GB is recommended. Using the full 262K context window can add up to 17.9 GB, bringing total usage to 43.7 GB. 1 GPU can run it, including NVIDIA GeForce RTX 5090. 98 devices with unified memory can run Qwen3 42B A3B 2507 Thinking Abliterated Uncensored TOTAL RECALL v2 Medium MASTER CODER, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB), Mac Studio M5 Max (36 GB).
20 GB
Q4_K_M · best fit Q2_K (19.4 GB)
1 GPU
RX 7900 XT~57Simulate output on RX 7900 XT32–40 GB
fits · 5.2 GB free · up to 64K ctx
1 GPU · 3 Mac classes · 2 device families
RTX 5090~166Simulate output on RTX 5090M5 Max36 GB~46Simulate output on Mac Studio M5 Max (36 GB)M532 GB~26Simulate output on MacBook Air 13" M5 (32 GB)Orin~42Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~27Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
Alternatives to Qwen3 42B A3B 2507 Thinking Abliterated Uncensored TOTAL RECALL v2 Medium MASTER CODER
Qwen 3 family · 77 models →Similar models
3| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Phi 3.5 MoE InstructMicrosoft | — | 23.9 GB | ~141 | same size |
| Falcon 40B InstructTII UAE | — | 23.8 GB | ~33 | same size |
| IQuest Coder V1 40B Loop InstructIQuestLab | — | 85.3 GB | —needs > 24 GB | same size |
Specifications
- Publisher
- DavidAU
- Family
- Qwen 3
- Parameters
- 42.4B
- Active parameters
- 4.4B per token
- Architecture
- Qwen3MoeForCausalLM
- Context Length
- 262,144 tokens
- Vocabulary Size
- 151,936
- Release Date
- 2025-08-21
- License
- Apache 2.0
Frequently Asked Questions
How much VRAM does Qwen3 42B A3B 2507 Thinking Abliterated Uncensored TOTAL RECALL v2 Medium MASTER CODER need?
Qwen3 42B A3B 2507 Thinking Abliterated Uncensored TOTAL RECALL v2 Medium MASTER CODER requires 25.9 GB of VRAM at Q4_K_M, or 85.2 GB at BF16. Full 262K context adds up to 17.9 GB (43.7 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 42.4B × 4.8 bits ÷ 8 = 25.4 GB
KV Cache + Overhead ≈ 0.5 GB (at 2K context + ~0.3 GB framework)
Fit ratings and hardware model lists check this model with room for a 16K-token context, which needs a little more memory.
KV Cache + Overhead ≈ 18.3 GB (at full 262K context)
VRAM usage by quantization
Can NVIDIA GeForce RTX 4090 run Qwen3 42B A3B 2507 Thinking Abliterated Uncensored TOTAL RECALL v2 Medium MASTER CODER?
Yes, at Q3_K_M (21.1 GB) or lower. Higher quantizations like Q4_K_M (25.9 GB) exceed the NVIDIA GeForce RTX 4090's 24 GB.
What's the best quantization for Qwen3 42B A3B 2507 Thinking Abliterated Uncensored TOTAL RECALL v2 Medium MASTER CODER?
For Qwen3 42B A3B 2507 Thinking Abliterated Uncensored TOTAL RECALL v2 Medium MASTER CODER, Q4_K_M (25.9 GB) offers the best balance of quality and VRAM usage. Q5_K_M (30.6 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 18.4 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run Qwen3 42B A3B 2507 Thinking Abliterated Uncensored TOTAL RECALL v2 Medium MASTER CODER on a Mac?
Yes — Mac Mini M5 Pro (24 GB) and 106 other Macs can run Qwen3 42B A3B 2507 Thinking Abliterated Uncensored TOTAL RECALL v2 Medium MASTER CODER, the smallest at Q2_K. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 25.9 GB of usable unified memory (RAM minus macOS overhead), from Mac Studio M1 Max (32 GB).
Can I run Qwen3 42B A3B 2507 Thinking Abliterated Uncensored TOTAL RECALL v2 Medium MASTER CODER locally?
Yes — Qwen3 42B A3B 2507 Thinking Abliterated Uncensored TOTAL RECALL v2 Medium MASTER CODER can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Qwen3 42B A3B 2507 Thinking Abliterated Uncensored TOTAL RECALL v2 Medium MASTER CODER?
At Q4_K_M, Qwen3 42B A3B 2507 Thinking Abliterated Uncensored TOTAL RECALL v2 Medium MASTER CODER can reach ~72 tok/s on AMD Instinct MI350X. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = 1000 ÷ (active GB ÷ (bandwidth GB/s × efficiency) × 1000 + layers × routing ms)
Mixture-of-Experts: just the active experts get read for each token, along with a fixed routing cost per layer.
Example: NVIDIA B200 → 2.7 GB active ÷ (8000 × 0.65) = 0.52 ms, plus 67 layers × 0.055 ms = 3.69 ms, so 1000 ÷ 4.21 ms = ~238 tok/s
Estimated speed at Q4_K_M (25.9 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of Qwen3 42B A3B 2507 Thinking Abliterated Uncensored TOTAL RECALL v2 Medium MASTER CODER?
At Q4_K_M, the download is about 25.42 GB. The full-precision BF16 version is 84.74 GB. The smallest option (Q2_K) is 18.01 GB.
Which GPUs can run Qwen3 42B A3B 2507 Thinking Abliterated Uncensored TOTAL RECALL v2 Medium MASTER CODER?
1 consumer GPU can run Qwen3 42B A3B 2507 Thinking Abliterated Uncensored TOTAL RECALL v2 Medium MASTER CODER at Q4_K_M. Top options include NVIDIA GeForce RTX 5090.
Which devices can run Qwen3 42B A3B 2507 Thinking Abliterated Uncensored TOTAL RECALL v2 Medium MASTER CODER?
98 devices with unified memory can run Qwen3 42B A3B 2507 Thinking Abliterated Uncensored TOTAL RECALL v2 Medium MASTER CODER at Q4_K_M (25.9 GB), including ASUS Ascent GX10, Beelink GTR9 Pro (Ryzen AI Max+ 395, 128 GB), Framework Desktop (Ryzen AI Max+ 395, 128 GB), GMKtec EVO-X2 (Ryzen AI Max+ 395, 128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.