Hack Day Starter
Registry Version:v2026.10.04.1Verified:October 4, 2026 13:00 UTC

Verified Ollama Model Registry

An authoritative reference catalog of open-weight models evaluated for local laptop and workstation execution. Every entry is source-verified against official Ollama library manifests, with explicit distinctions between observed source facts, vendor publications, and empirical heuristics.

Tracked Entries
23
Total catalog records
Current Generation
20
Active primary candidates
Verified Records
21
Primary source agreement
Eligible Recommender
20
Passes local safety gates

Data Fidelity & Lifecycle Taxonomy

1. Source-Observed Data

Extracted directly from official Ollama library HTTP endpoints: exact manifest byte counts, short sha256 digests, quantization types, and context token ceilings.

2. Officially Documented Values

Minimum or recommended system RAM published in official foundation model documentation (e.g. Google DeepMind for Gemma, Alibaba for Qwen).

3. Hack Day Starter Estimates

Empirical sizing calculations accounting for weight size, KV cache expansion, and a 1.5 GB OS safety margin to prevent swapping during live demonstrations.

Current Generation Models (20)

Active primary candidates evaluated for local deployment

Gemma 4 e4b

GoogleQ4_K_MCurrent
ollama pull gemma4:e4b
License:Gemma Terms of Use

Google's 4B multimodal model with native tool-calling, vision inputs, and configurable reasoning tokens.

Source Manifest SizeOllama listed: 6.6GB
Context Token Window128k tokens(131,072 tokens)
Memory Requirement
8 GB System RAMOfficial Vendor Guidance
Verified Capabilities:✓ Tool-calling✓ Multimodal VisionReasoning tokens
Ollama PageModel Card

Gemma 4 12B

GoogleQ4_K_MCurrent
ollama pull gemma4:12b
License:Gemma Terms of Use

Flagship local model from Google offering high reasoning, multimodal perception, and 256k context.

Source Manifest SizeOllama listed: 7.7GB
Context Token Window256k tokens(262,144 tokens)
Memory Requirement
16 GB System RAMOfficial Vendor Guidance
Verified Capabilities:✓ Tool-calling✓ Multimodal VisionReasoning tokens
Ollama PageModel Card

Qwen 3.5 4B

Alibaba QwenQ4_K_MCurrent
ollama pull qwen3.5:4b
License:Apache-2.0

Compact Qwen 3.5 release featuring hybrid thinking tokens, strong code logic, vision input, and 256k context.

Source Manifest SizeOllama listed: 3.4GB
Context Token Window256k tokens(262,144 tokens)
Memory Requirement
~6 GB ComfortStarter Empirical Estimate
Verified Capabilities:✓ Tool-calling✓ Multimodal VisionReasoning tokens
Ollama PageModel Card

Qwen 3.5 9B

Alibaba QwenQ4_K_MCurrent
ollama pull qwen3.5:9b
License:Apache-2.0

Balanced 9B model excelling at complex multi-step reasoning, coding tasks, and 256k context window.

Source Manifest SizeOllama listed: 6.6GB
Context Token Window256k tokens(262,144 tokens)
Memory Requirement
~10 GB ComfortStarter Empirical Estimate
Verified Capabilities:✓ Tool-calling✓ Multimodal VisionReasoning tokens
Ollama PageModel Card

Qwen 3.6 27B

Alibaba QwenQ4_K_MCurrent
ollama pull qwen3.6:27b
License:Apache-2.0

Dense 27B model from Qwen 3.6 series with extended 256k context window and multimodal vision.

Source Manifest SizeOllama listed: 18GB
Context Token Window256k tokens(262,144 tokens)
Memory Requirement
~24 GB ComfortStarter Empirical Estimate
Verified Capabilities:✓ Tool-calling✓ Multimodal VisionReasoning tokens
Ollama PageModel Card

Qwen 3.6 35B

Alibaba QwenQ4_K_MCurrent
ollama pull qwen3.6:35b
License:Apache-2.0

35B parameter model with 23 GB download footprint and full 256k context for high-memory setups.

Source Manifest SizeOllama listed: 23GB
Context Token Window256k tokens(262,144 tokens)
Memory Requirement
~30 GB ComfortStarter Empirical Estimate
Verified Capabilities:✓ Tool-calling✓ Multimodal VisionReasoning tokens
Ollama PageModel Card

Qwen 3.8 27B

Alibaba QwenQ4_K_MCurrent
ollama pull qwen3.8:27b
License:Apache-2.0

Qwen 3.8 release (August 2026) with refined reasoning architecture, 256k context, and multimodal vision.

Source Manifest SizeOllama listed: 18GB
Context Token Window256k tokens(262,144 tokens)
Memory Requirement
~24 GB ComfortStarter Empirical Estimate
Verified Capabilities:✓ Tool-calling✓ Multimodal VisionReasoning tokens
Ollama PageModel Card

Qwen 3 8B

Alibaba QwenQ4_K_MCurrent
ollama pull qwen3:8b
License:Apache-2.0

Standard dense 8B model with fast latency and proven tool calling for intermediate laptops.

Source Manifest SizeOllama listed: 5.2GB
Context Token Window40k tokens(40,960 tokens)
Memory Requirement
~8 GB ComfortStarter Empirical Estimate
Verified Capabilities:✓ Tool-calling✕ Text only
Ollama PageModel Card

GPT-OSS 20B

Open-Weight CommunityQ4_K_MCurrent
ollama pull gpt-oss:20b
License:Apache-2.0

Open-weight architecture with 21B total parameters (3.6B active) and 14 GB Ollama download.

Source Manifest SizeOllama listed: 14GB
Context Token Window128k tokens(131,072 tokens)
Memory Requirement
~18 GB ComfortStarter Empirical Estimate
Verified Capabilities:✓ Tool-calling✕ Text onlyReasoning tokens
Ollama PageModel Card

Phi-4 Mini

MicrosoftQ4_K_MCurrent
ollama pull phi4-mini
License:MIT

Microsoft's 3.8B model with strong synthetic math, reasoning, and tool support for lightweight machines.

Source Manifest SizeOllama listed: 2.5GB
Context Token Window128k tokens(131,072 tokens)
Memory Requirement
~4 GB ComfortStarter Empirical Estimate
Verified Capabilities:✓ Tool-calling✕ Text only
Ollama PageModel Card

Nemotron 3 Nano 4B

NVIDIAQ4_K_MCurrent
ollama pull nemotron-3-nano:4b
License:NVIDIA Open Model License

NVIDIA's 4B edge model with 2.8 GB footprint and 256k context engineered for low-latency tool calling.

Source Manifest SizeOllama listed: 2.8GB
Context Token Window256k tokens(262,144 tokens)
Memory Requirement
~5 GB ComfortStarter Empirical Estimate
Verified Capabilities:✓ Tool-calling✕ Text only
Ollama PageModel Card

Qwen 3 Coder Next

Alibaba QwenQ4_K_MCurrent
ollama pull qwen3-coder-next:latest
License:Apache-2.0

80B sparse MoE coding model with 3B active parameters, 52 GB footprint, and 256k context for high-end workstations.

Source Manifest SizeOllama listed: 52GB
Context Token Window256k tokens(262,144 tokens)
Memory Requirement
~60 GB ComfortStarter Empirical Estimate
Verified Capabilities:✓ Tool-calling✕ Text onlyReasoning tokens
Ollama PageModel Card

DeepSeek R1 1.5B

DeepSeekQ4_K_MCurrent
ollama pull deepseek-r1:1.5b
License:MIT

Distilled 1.5B reasoning model with chain-of-thought outputs. Runs fast on CPU; does not support native function calling.

Source Manifest SizeOllama listed: 1.1GB
Context Token Window128k tokens(131,072 tokens)
Memory Requirement
~3 GB ComfortStarter Empirical Estimate
Verified Capabilities:✕ No tools✕ Text onlyReasoning tokens
Ollama PageModel Card

GLM-4 9B (Unverified Candidate)

Zhipu AIQ4_0Current
ollama pull glm4:9b
License:GLM-4 License

Candidate model entry pending full verification audit.

Source Manifest SizeOllama listed: 5.5GB
Context Token Window128k tokens(131,072 tokens)
Memory Requirement
~9 GB ComfortStarter Empirical Estimate
Verified Capabilities:✓ Tool-calling✕ Text only
Ollama PageModel Card

Llama 3.2 3B

MetaQ4_K_MCurrent✓ Runtime Verified
ollama pull llama3.2:3b
License:Llama 3.2 Community License

Meta's 3B lightweight model. Runtime verified live through Hack Day Starter Chat + Tool-calling Agent.

Source Manifest SizeOllama listed: 2.0GB
Context Token Window128k tokens(131,072 tokens)
Memory Requirement
~4 GB ComfortStarter Empirical Estimate
Verified Capabilities:✓ Tool-calling✕ Text only
Ollama PageModel Card

Llama 3.1 8B

MetaQ4_K_MCurrent✓ Runtime Verified
ollama pull llama3.1:8b
License:Llama 3.1 Community License

Meta's foundational 8B model with 128k context. Runtime verified live through Hack Day Starter Chat + Tool-calling Agent.

Source Manifest SizeOllama listed: 4.9GB
Context Token Window128k tokens(131,072 tokens)
Memory Requirement
~8 GB ComfortStarter Empirical Estimate
Verified Capabilities:✓ Tool-calling✕ Text only
Ollama PageModel Card

Qwen 2.5 Coder 7B

Alibaba QwenQ4_K_MCurrent⏳ Runtime Pending
ollama pull qwen2.5-coder:7b
License:Apache-2.0

Alibaba's dedicated 7B code generation model with 32k context. Source/capability verified; runtime verification pending.

Source Manifest SizeOllama listed: 4.7GB
Context Token Window32k tokens(32,768 tokens)
Memory Requirement
~8 GB ComfortStarter Empirical Estimate
Verified Capabilities:✓ Tool-calling✕ Text only
Ollama PageModel Card

Mistral NeMo 12B

Mistral AI / NVIDIAQ4_K_MCurrent⏳ Runtime Pending
ollama pull mistral-nemo:12b
License:Apache-2.0

12B multilingual model with Tekken tokenizer and 1000k context. Source/capability verified; runtime verification pending.

Source Manifest SizeOllama listed: 7.1GB
Context Token Window1000k tokens(1,024,000 tokens)
Memory Requirement
~11 GB ComfortStarter Empirical Estimate
Verified Capabilities:✓ Tool-calling✕ Text only
Ollama PageModel Card

LFM 2.5 8B

Liquid AIQ4_K_MCurrent⏳ Runtime Pending
ollama pull lfm2.5:8b
License:Liquid AI Model License

Liquid Foundation Model with reasoning tokens and 125k context. Source/capability verified; runtime verification pending.

Source Manifest SizeOllama listed: 5.2GB
Context Token Window125k tokens(128,000 tokens)
Memory Requirement
~9 GB ComfortStarter Empirical Estimate
Verified Capabilities:✓ Tool-calling✕ Text onlyReasoning tokens
Ollama PageModel Card

Devstral Small 2

Mistral AI CommunityQ4_K_MCurrent⏳ Runtime Pending
ollama pull devstral-small-2:latest
License:Apache-2.0

24B multimodal code agent model with 384k context. Source/capability verified; runtime verification pending.

Source Manifest SizeOllama listed: 15GB
Context Token Window384k tokens(393,216 tokens)
Memory Requirement
~19 GB ComfortStarter Empirical Estimate
Verified Capabilities:✓ Tool-calling✓ Multimodal Vision
Ollama PageModel Card

Legacy Models (2)

Older model generations superseded by newer family revisions

Gemma 3 12B

Legacy

Legacy Gemma 3 release superseded by Gemma 4. Kept in registry for historical compatibility.

ollama pull gemma3:12b
Superseded by newer architecture releases. Still functional in local Ollama runtimes.

Mistral 7B v0.1 (Stale Metadata)

Legacy

Legacy Mistral model with outdated verification metadata.

ollama pull mistral:v0.1
Superseded by newer architecture releases. Still functional in local Ollama runtimes.

Retired & Unverified Candidates (3)

Tracked for catalog integrity; strictly barred from active recommendations
Llama 2 7B (Retired)(llama2:7b)Retired

Deprecated 2023 release. Marked as retired in registry.

Safety Gate: Excluded from recommendations
Mistral 7B v0.1 (Stale Metadata)(mistral:v0.1)Metadata Stale

Legacy Mistral model with outdated verification metadata.

Safety Gate: Excluded from recommendations
GLM-4 9B (Unverified Candidate)(glm4:9b)Unverified Candidate

Candidate model entry pending full verification audit.

Safety Gate: Excluded from recommendations
← Back to Hardware Profiler