AI Model Database

Model Fields

Model pages should track task fit, model family, parameter size, quantization, context length, license, release date, runtime support, hardware requirement, privacy boundary, and known limitations. A model that benchmarks well may still be wrong for a local private workflow.

Review Cadence

Model releases, licenses, context windows, and runtimes change quickly. Preserve version dates and recheck any recommendation before using it for buying hardware or processing sensitive data.

Last reviewed May 24, 2026.

Source Basis

AI pages should cite model cards, project repositories, release notes, hardware vendor specifications, driver/runtime documentation, and D-Central infrastructure experience. Benchmark claims should preserve model version, quantization, context length, hardware, driver, runtime, and test date.

ASIC miners do not run LLM workloads. D-Central connects AI to its practical infrastructure domain through power, heat, privacy, local compute, maintenance, and hardware operations.

Reviewer

Reviewed by D-Central editorial staff with a Bitcoin infrastructure, privacy, hardware, and operations lens. Sensitive data, private keys, customer records, and production secrets should not be loaded into experimental AI stacks.

Freshness Policy

Model releases, licenses, GPU prices, driver support, inference runtimes, and leaderboard results change quickly. AI pages should preserve the model and hardware version used, identify stale benchmarks, and separate local privacy guidance from performance claims.

Last reviewed May 24, 2026. D-Central editorial and repair intake, Laval, Quebec.

40 local AI model profiles for sovereign compute planning: model families, context windows, quantization notes, hardware fit, and deployment considerations.

Model Profiles

Qwen 3

Alibaba's May 2025 release — first open family with hybrid reasoning (toggle-able chain of thought), Apache 2.0 across all sizes.

View Model

Llama 4 (Scout/Maverick)

Meta's April 2025 MoE-and-multimodal release, headlined by Scout's 10M-token context window and the pre-announced Behemoth frontier model.

View Model

Gemma 3

Google DeepMind's March 2025 Gemma family — vision-capable (4B+), 128K context, with official quantization-aware 4-bit variants.

View Model

Mistral Small 3

Mistral AI's January 2025 24B model — Apache 2.0, competitive with Llama 3.3 70B, fits on a single 24GB GPU.

View Model

DeepSeek R1

DeepSeek's January 2025 reasoning model — frontier chain-of-thought quality, plus six MIT-licensed distills from 1.5B to 70B.

View Model

DeepSeek V3

DeepSeek's December 2024 frontier-scale MoE — 671B total, 37B active, trained for ~$5.6M in compute.

View Model

Phi-4

Microsoft Research's December 2024 Phi-4 — a 14B dense MIT-licensed model punching well above its weight on math and reasoning.

View Model

Llama 3.3

A single 70B model released December 2024, closing most of the gap to Llama 3.1 405B through improved post-training alone.

View Model

Stable Diffusion 3.5

Stability AI's October 2024 MMDiT flagship — 2B (Medium) and 8B (Large) variants with dramatically improved prompt adherence over SDXL.

View Model

Llama 3.2

Meta's September 2024 Llama release added edge sizes (1B/3B) and the first open-weight Llama vision models (11B/90B).

View Model

Qwen 2.5

Alibaba's September 2024 Qwen family spans 0.5B to 72B, plus coding and math specialists — mostly Apache 2.0.

View Model

FLUX.1 schnell

Black Forest Labs' August 2024 Apache 2.0 FLUX variant — 12B distilled to 1-4 steps for fast, commercially-open image generation.

View Model

FLUX.1 dev

Black Forest Labs' August 2024 flagship — 12B rectified-flow transformer that set the new quality bar for open image generation.

View Model

Llama 3.1

Meta's flagship 2024 open-weight LLM family — 8B, 70B, and 405B parameters with 128K context. The 405B was the first open-weight model at true frontier scale.

View Model

Gemma 2

Google DeepMind's June 2024 lightweight open model family — 2B, 9B, and 27B with interleaved local/global attention.

View Model

Command R+

Cohere's April 2024 RAG-native flagship — 104B dense, first-class grounded citation and tool use, CC-BY-NC 4.0.

View Model

Mixtral 8x7B

Mistral AI's December 2023 mixture-of-experts model — 8 experts, 2 active per token, Apache 2.0, ran at Llama-13B speed with Llama-70B quality.

View Model

Whisper Large v3

OpenAI's November 2023 open ASR model — 1.55B params, MIT-licensed, the open reference for multilingual speech-to-text.

View Model

Mistral 7B

Mistral AI's September 2023 debut — a 7B Apache-2.0 model that popularized Grouped-Query and Sliding Window Attention.

View Model

Stable Diffusion XL

Stability AI's July 2023 SDXL — ~3.5B params, 1024x1024 native, CreativeML-OpenRAIL-M license, backbone of the open image generation ecosystem.

View Model

Methodology

Model data is maintained as a planning reference for local inference. Always verify license terms, model card updates, context length, quantization compatibility, and hardware memory requirements before deployment.

Editorial review and limitations

Reviewed by D-Central's mining hardware and ASIC repair editorial team for practical accuracy, buyer risk, repair context, and operational assumptions. Verify current hardware price, stock, network difficulty, BTC price, power rate, shipping, tax, firmware, and device condition before buying, hosting, repairing, or retiring mining hardware.

Last reviewed May 24, 2026. D-Central, Laval, Quebec.

Related products, repair, and setup paths

Last reviewed May 25, 2026. Added to route educational and product traffic toward exact commercial next steps.