DavidAU/Qwen3-Coder-53B-A3B-Instruct-TOTAL-RECALL-v2-MASTER-CODER-L

Text Generation • 53B • Updated Aug 3 • 47 • 9

Note 128 experts (MOE) - Mixture of experts. All experts are coders. 256K context ; using Brainstorm 40x to enhance performance. Non-Thinking model (you can activate this using a system prompt) Links to GGUFs on this page.

DavidAU/Qwen3-Coder-42B-A3B-Instruct-TOTAL-RECALL-MASTER-CODER-M

Text Generation • 42B • Updated Aug 2 • 39 • 3

Note 128 experts (MOE) - Mixture of experts. All experts are coders. 256K context ; using Brainstorm 20x to enhance performance. Non-Thinking model (you can activate this using a system prompt) Links to GGUFs on this page.

DavidAU/Qwen3-Coder-42B-A3B-Instruct-TOTAL-RECALL-MASTER-CODER-M-512k-ctx

Text Generation • 42B • Updated Aug 3 • 37 • 2

Note 128 experts (MOE) - Mixture of experts. All experts are coders. 512K context ; using Brainstorm 20x to enhance performance. Non-Thinking model (you can activate this using a system prompt) Links to GGUFs on this page. Special note: Even you do not need the context, try this model as context changes will change generation.

DavidAU/Qwen3-Coder-42B-A3B-Instruct-TOTAL-RECALL-MASTER-CODER-M-1million-ctx

Text Generation • 42B • Updated Aug 3 • 10 • 6

Note 128 experts (MOE) - Mixture of experts. All experts are coders. 1 million context ; using Brainstorm 20x to enhance performance. Non-Thinking model (you can activate this using a system prompt) Links to GGUFs on this page. Special note: Even you do not need the context, try this model as context changes will change generation.

DavidAU/Qwen3-53B-A3B-2507-THINKING-TOTAL-RECALL-v2-MASTER-CODER

Text Generation • 53B • Updated Aug 2 • 13 • 8

Note 128 experts (MOE) - Mixture of experts. 256K context ; using Brainstorm 40x to enhance performance. Non-Thinking model (you can activate this using a system prompt) Links to GGUFs on this page.

DavidAU/Qwen3-53B-A3B-2507-TOTAL-RECALL-v2-MASTER-CODER

Text Generation • 53B • Updated Aug 2 • 14 • 8

Note 128 experts (MOE) - Mixture of experts. 256K context ; using Brainstorm 40x to enhance performance. Links to GGUFs on this page. Non-thinking model => STRAIGHT to coding.

DavidAU/Qwen3-42B-A3B-2507-Thinking-TOTAL-RECALL-v2-Medium-MASTER-CODER

Text Generation • 42B • Updated Aug 2 • 214 • 4

Note 128 experts (MOE) - Mixture of experts. 256K context ; using Brainstorm 20x to enhance performance. Links to GGUFs on this page. Enhanced Thinking model => Smarter thinking, fewer tokens, better code.

DavidAU/Qwen3-42B-A3B-2507-TOTAL-RECALL-v2-Medium-MASTER-CODER

Text Generation • 42B • Updated Aug 2 • 6 • 3

Note 128 experts (MOE) - Mixture of experts. 256K context ; using Brainstorm 20x to enhance performance. Links to GGUFs on this page. Non-thinking model => STRAIGHT to coding.

DavidAU/Qwen3-53B-A3B-TOTAL-RECALL-MASTER-CODER-v1.4

Text Generation • 53B • Updated Jul 21 • 5 • 3

Note 128 experts MOE Model. Links to quants on this page -> GGUF, GGUF Imatrix and others are here. Uses Brainstorm adapter (40x) by DavidAU to extend model function/performance.

DavidAU/Qwen2.5-2X7B-Coder-Soar-qwen-Coder-Instruct-OlympicCoder-19B

Text Generation • 19B • Updated Jul 31 • 2 • 2

Note Specialized 2 model with MOE with additional shared expert. Links to quants on this page -> GGUF, GGUF Imatrix and others are here.

DavidAU/Mistral-Magistral-Devstral-Instruct-FUSED-CODER-Reasoning-36B

Text Generation • 36B • Updated Jul 31 • 15 • 4

Note Newest Devstral version (1.1), with even better coding abilities. Links to quants on this page -> GGUF, GGUF Imatrix and others are here. Context at 128k. This is a fused model with 62 layers, 561 tensors. Short thinking blocks -> then straight to coding.

DavidAU/Mistral-Devstral-2507-CODER-Brainstorm40x-44B

Text Generation • 44B • Updated Jul 31 • 4 • 2

Note Newest Devstral version, with even better coding abilities. Links to quants on this page -> GGUF, GGUF Imatrix and others are here. Context at 128k. Uses Brainstorm adapter by DavidAU to extend model function/performance.

DavidAU/Mistral-Devstral-2505-CODER-Brainstorm40x-44B

Text Generation • 44B • Updated Jul 31 • 113 • 2

Note Links to quants on this page -> GGUF, GGUF Imatrix and others are here. Context at 128k. Uses Brainstorm adapter by DavidAU to extend model function/performance.

DavidAU/Qwen3-Esper3-Reasoning-CODER-Instruct-21B-Brainstorm20x

Text Generation • 21B • Updated Jul 31 • 7 • 2

Note Links to quants on this page -> GGUF, GGUF Imatrix and others are here. Uses Brainstorm adapter by DavidAU to extend model function/performance.

DavidAU/Qwen2.5-Microsoft-NextCoder-Brainstorm20x-128k-ctx-12B

Text Generation • 12B • Updated Jul 31 • 3 • 4

Note Links to quants on this page -> GGUF, GGUF Imatrix and others are here. Context at 128k. Uses Brainstorm adapter by DavidAU to extend model function/performance.

DavidAU/Qwen3-Esper3-Reasoning-CODER-Instruct-12B-Brainstorm20x

Text Generation • 12B • Updated Jul 31 • 32 • 2

Note Uses Brainstorm adapter by DavidAU to extend model function/performance. Links to quants on this page -> GGUF, GGUF Imatrix and others are here.

DavidAU/Qwen3-Jan-Nano-128k-6B-Brainstorm20x

Text Generation • 6B • Updated Jul 31 • 22 • 4

Note Uses Brainstorm adapter by DavidAU to extend model function/performance. Links to quants on this page -> GGUF, GGUF Imatrix and others are here. This is a general use AND coder/programming model.

DavidAU/Qwen3-Blitzar-Coder-F1-6B-Brainstorm20x

Text Generation • 6B • Updated Jul 31 • 3 • 3

Note Uses Brainstorm adapter by DavidAU to extend model function/performance. Links to quants on this page -> GGUF, GGUF Imatrix and others are here.

DavidAU/Qwen3-Polaris-Preview-128k-6B-Brainstorm20x

Text Generation • 6B • Updated Jul 31 • 12 • 2

Note Uses Brainstorm adapter by DavidAU to extend model function/performance. Links to quants on this page -> GGUF, GGUF Imatrix and others are here. This is a general use AND coder/programming model.

DavidAU/Qwen3-Instruct-6B-Brainstorm20x-128k-ctx

Text Generation • 6B • Updated Jul 31 • 6 • 1

Note Uses Brainstorm adapter by DavidAU to extend model function/performance. Links to quants on this page -> GGUF, GGUF Imatrix and others are here. This is a general use AND coder/programming model. 128 k context.

DavidAU/Qwen3-Code-Reasoning-Instruct-6B-Brainstorm20x

Text Generation • 6B • Updated Jul 31 • 48 • 1

Note Uses Brainstorm adapter by DavidAU to extend model function/performance. Links to quants on this page -> GGUF, GGUF Imatrix and others are here. This is a general use AND coder/programming model.

DavidAU/Qwen3-Esper3-Reasoning-Instruct-6B-Brainstorm20x-Enhanced-E32-128k-ctx

Text Generation • 6B • Updated Jul 31 • 32

Note Uses Brainstorm adapter by DavidAU to extend model function/performance. Links to quants on this page -> GGUF, GGUF Imatrix and others are here. Float 32 enhanced, and 128k context.

DavidAU/Qwen3-Zero-Coder-Reasoning-0.8B-NEO-EX-GGUF

Text Generation • 0.8B • Updated Jul 28 • 11.9k • 15

Note Uses NEO Imatrix dataset (by DavidAU) to augment model performance. 40k context. Good for draft, simple, or code blocks - including complex. Model has full thinking/reasoning too. Model is fused together from 2 coder models

DavidAU/Qwen3-Zero-Coder-Reasoning-V2-0.8B-NEO-EX-GGUF

Text Generation • 0.8B • Updated Jul 28 • 7.03k • 13

Note Links to quants on this page -> GGUF, GGUF Imatrix and others are here. 40k context. Good for draft, simple, or code blocks - including complex. Model has full thinking/reasoning too. Stronger than V1. Model is fused together from 2 coder models.

DavidAU/Qwen3-Zero-Coder-Reasoning-0.8B

Text Generation • 0.8B • Updated Jul 31 • 4 • 2

Note Links to quants on this page -> GGUF, GGUF Imatrix and others are here. 40k context. Good for draft, simple, or code blocks - including complex. Model has full thinking/reasoning too. Model is fused together from 2 coder models

DavidAU/Qwen3-Zero-Coder-Reasoning-V2-0.8B

Text Generation • 0.8B • Updated Jul 31 • 96

Note Links to quants on this page -> GGUF, GGUF Imatrix and others are here. 40k context. Good for draft, simple, or code blocks - including complex. Model has full thinking/reasoning too. Stronger than V1. Model is fused together from 2 coder models.