-
thoxnova-12b-agent
THOX.ai 12B tool-calling agent configuration. LoRA fine-tune with tool use. Fully public. Apache-2.0.
vision tools thinking507 Pulls 2 Tags Updated 2 days ago
-
thoxheretic
Community uncensored (abliterated) Qwen3.5-9B creative-writing merge, redistributed by THOX as a Q4_K_M GGUF with a paired vision projector. Not THOX-trained and not safety-tuned; THOX evaluation pending. Apache-2.0.
vision 9b223 Pulls 1 Tag Updated 2 days ago
-
thoxmythos
Community uncensored Qwen3.5-9B redistribution (Qwythos 9B Mythos build, Empero AI), published by THOX as a runnable Ollama model, Q5_K_M, 6.6 GB. Not THOX-trained; no THOX evaluation published. Apache-2.0.
77 Pulls 3 Tags Updated 2 days ago
-
thoxedge-0.8b
The ThoxEdge flagship: a merged 0.8B build that answers from canonical THOX facts and declines what THOX has not published. GATE49 46/49 (development set), Q4_K_M GGUF.
68 Pulls 2 Tags Updated yesterday
-
thoxforge-7b
THOX.ai coding and reasoning assistant. 7B LoRA fine-tune, Q4_K_M, tuned for Rust / TypeScript / Python. Fully public. Apache-2.0.
tools67 Pulls 2 Tags Updated 2 days ago
-
thoxmini-3b
ThoxMini Assistant (LoRA Adapter + GGUF) — a THOX assistant model for the ThoxMini tier: a LoRA fine-tune (rank 16) on a THOX-authored corpus of 1,565 instruction examples, published as an adapter with a bundled Q4_K_M GGUF for direct use
tools67 Pulls 2 Tags Updated 20 hours ago
-
thoxnova-12b-core
THOX.ai 12B local assistant configuration. LoRA fine-tune with Q4_K_M + Q5_K_M GGUF builds. Fully public. Apache-2.0.
vision tools thinking65 Pulls 2 Tags Updated 2 days ago
-
thoxair-0.5b
The ThoxAir flagship: canonical THOX facts and house behaviour, fine-tuned into a 0.5B model that runs anywhere. GATE49 47/49 (development set), merged weights, Q4_K_M GGUF.
tools51 Pulls 2 Tags Updated 3 days ago
-
thoxclip-qwen3-0.6b
The on-device assistant for ThoxClip. GATE49 22/49 baseline (development set). Grounding required: serve thinking-off with the grounded THOX device-facts system prompt.
tools thinking47 Pulls 2 Tags Updated 3 days ago
-
thoxopen-vision-heretic
Minimal-refusal vision model: THOX Open profile on abliterated Gemma 4 E2B (QAT Q4), projector included. 120.44 tok/s, 1.86 GB footprint on a GPU host at 8,192 ctx; grounding 8/8, trap declined. Upstream weights. Apache-2.0.
vision tools thinking44 Pulls 1 Tag Updated yesterday
-
thoxflux-350m
ThoxClip Edge Assistant — an on-device assistant model for the ThoxClip tier: a LoRA fine-tune (rank 32) on a THOX-authored corpus of 982 chat, tool-use, and factuality examples, merged into the base weights. GATE49 20/49 (development set)
tools41 Pulls 3 Tags Updated 2 days ago
-
thoxedge-0.5b
ThoxMini Edge Assistant — an on-device assistant model for the ThoxMini tier: a LoRA fine-tune (rank 32) on a THOX-authored corpus of 982 chat, tool-use, and factuality examples, merged into the base weights. GATE49 47/49 (development set)
tools40 Pulls 3 Tags Updated 4 days ago
-
thoxwave-8b
Mid-Size Chat LoRA — THOX's mid-size chat fine-tune for the THOX stack: a LoRA adapter (rank 8, context 8192, ChatML template) with a bundled Q4_K_M GGUF (4.92 GB) Safetensors weights · GGUF: Q4_K_M
39 Pulls 1 Tag Updated 3 months ago
-
thoxopen-8b-heretic
THOX's own abliteration of Qwen3-8B (Heretic): refusals 13/100 vs stock 100/100, KL 0.0482, capability 8/8. GGUF Q5_K_M 5.85 GB. THOX abliteration, not THOX-trained. Apache-2.0.
38 Pulls 3 Tags Updated yesterday
-
thoxmini-phi4mini
The ThoxMini flagship assistant, and the host payload ThoxBeam serves to ThoxKey. GATE49 47/49 (development set); baseline established, not promoted. Serve with the grounding prompt.
tools36 Pulls 2 Tags Updated 3 days ago
-
thoxnano-135m
ThoxKey Tier Model — the smallest THOX device model, built for the ThoxKey tier: a LoRA fine-tune (rank 32) on a THOX-authored corpus of 982 chat, tool-use, and factuality examples, merged into the base weights. GATE49 23/49 (development set)
tools36 Pulls 3 Tags Updated 2 days ago
-
thoxnano-360m
ThoxClip Edge Assistant — an on-device assistant model for the ThoxClip tier: a LoRA fine-tune (rank 32) on a THOX-authored corpus of 982 chat, tool-use, and factuality examples, merged into the base weights. GATE49 23/49 (development set)
tools35 Pulls 3 Tags Updated 2 days ago
-
thoxroute
THOX.ai model-routing classifier. Clean-room Qwen2.5-1.5B fine-tune that picks the best THOX model for a query. Infrastructure.
tools 1.5b34 Pulls 2 Tags Updated 2 days ago
-
thoxglobal-7b
THOX.ai multilingual assistant. 7B LoRA fine-tune, Q4_K_M, answers in the user's language. Fully public. Apache-2.0.
tools27 Pulls 1 Tag Updated 4 days ago
-
thoxmicro
THOX.ai from-scratch micro base model (ThoxMicro-125M, 123M parameters). A base and completion model, not an assistant; foundation for THOX device role models.
22 Pulls 2 Tags Updated 2 months ago
-
thoxwork-code-14b
Workgroup coding assistant on Qwen3-14B with THOX profiles and Knowledge Pack. 26.28 tok/s long-form, 10.49 GB footprint on a GPU host at 8,192 ctx; grounding 8/8, trap declined. Config-only, not THOX-trained. Apache-2.0.
tools thinking21 Pulls 1 Tag Updated yesterday
-
thoxmesh-32b
MeshStack dense baseline on Qwen3-32B with the THOX Knowledge Pack. 4.87 tok/s long-form, 22.61 GB footprint at 8,192 ctx; factory 3/3, grounding 8/8, trap declined. Config-only, not THOX-trained. Apache-2.0.
tools thinking19 Pulls 1 Tag Updated yesterday
-
thoxos-kernel-0.8b
ThoxOS kernel/OS triage student for ThoxAir and ThoxKey, distilled from ThoxOS-Kernel-27B. GATE-PASSED: GATE49 37/49, zero hard failures (development set). Q4_K_M GGUF.
18 Pulls 2 Tags Updated yesterday
-
thoxllm
THOX.ai from-scratch foundation model, 327M (v2 build). THOX-designed decoder, no external base model, Q4_K_M. Apache-2.0.
18 Pulls 1 Tag Updated 2 days ago
-
thoxos-kernel-27b
THOX's host-tier ThoxOS kernel/OS model and family teacher. Full-parameter fine-tune, GATE-PASSED under the adopted final-answer scope (development set). Open weights (D11). Q4_K_M GGUF.
16 Pulls 2 Tags Updated yesterday
-
thoxos-kernel-4b
ThoxOS kernel/OS resident student for ThoxMini and ThoxClip, distilled from ThoxOS-Kernel-27B. DOCUMENTED BASELINE: GATE49 44/49; safety-form bars held (see card). Q4_K_M GGUF.
15 Pulls 2 Tags Updated yesterday
-
thoxmini-125m-role
THOX.ai ThoxMini device-role model. Full fine-tune of the THOX-owned ThoxMicro-125M base; answers device commands on a paired host and defers the rest.
14 Pulls 1 Tag Updated 2 months ago
-
thox-pro-24b
A THOX configuration and distribution of Mistral AI's Devstral Small 2 24B, an agentic coding model, published for local software-engineering work. Apache-2.0.
vision tools14 Pulls 1 Tag Updated 10 months ago
-
thoxmesh-30b-a3b
THOX Mesh MoE model: Qwen3-30B-A3B packaged by THOX with the THOX Mesh system profile and THOX Knowledge Pack v0.1. Measured 11.1 tok/s long-form, 19.6 GB RAM. Configuration-derived; THOX did not train the weights. Apache-2.0.
tools thinking14 Pulls 1 Tag Updated 2 days ago
-
thox-sadie
THOX digital-human persona: Sadie Weaving, Chief Agentic Officer. 9.13 tok/s, 6.68 GB at 8,192 ctx. Configuration-derived from Qwen3-8B; not THOX-trained. Apache-2.0.
tools thinking13 Pulls 1 Tag Updated 2 days ago
-
thoxkey-route-0.5b
ThoxKey routing-proposal configuration. THOX-configured Qwen2.5-0.5B, not THOX-trained. 33.67 tok/s long-form, 0.59 GB RAM on CPU at 8,192 ctx; grounding 6/8, trap declined. Apache-2.0.
tools13 Pulls 1 Tag Updated 2 days ago
-
thoxkey-micro-0.5b
ThoxKey micro assistant (Qwen2.5-0.5B, THOX-Configured) — thoxKey micro assistant: short answers and lightweight assistance for the ThoxKey configuration 34.08 tok/s long-form (measured 2026-10-31, conditions under Evaluation) · Ollama Modelfile
tools13 Pulls 1 Tag Updated 2 days ago
-
thoxintel
ThoxIntel-27B: Qwen3.6-27B community-merge derivative staged by THOX (Q4_K_M, 19 GB, vision projector). THOX evaluation PENDING; weights not THOX-trained. Apache-2.0.
vision 27b12 Pulls 1 Tag Updated 2 days ago
-
thoxair-16m-role
ThoxAir Device-Role Model — the device-role conversational model for the ThoxAir persona, fine-tuned from a THOX model trained from scratch (no external base) Safetensors weights · GGUF: Q8_0, TQ1_0
12 Pulls 1 Tag Updated 2 months ago
-
thoxmesh-dense-32b
MeshStack routing assistant on dense Qwen3-32B with the THOX Knowledge Pack. 5.39 tok/s long-form, 22.61 GB footprint on a GPU host at 8,192 ctx; grounding 8/8, trap declined. Config-only, not THOX-trained. Apache-2.0.
tools thinking12 Pulls 1 Tag Updated yesterday
-
thoxopen-14b
THOX-configured Qwen3-14B laptop model (not THOX-trained). Measured 1.04 tok/s long-form, 10.3 GiB footprint at 8,192 ctx on cloud CPU; needs a 16 GB+ RAM host. Apache-2.0.
tools thinking12 Pulls 1 Tag Updated yesterday
-
thoxhuman-chat-8b
THOX-configured Qwen3-8B model for digital-human conversation, with the THOX agent profile and Knowledge Pack v0.1. 49.78 tok/s long-form, 6.31 GB footprint (GPU host, 2026-11-01). Config-derived; THOX did not train these weights. Apache-2.0.
tools thinking11 Pulls 1 Tag Updated yesterday
-
thox-atlas
THOX digital-human persona: Atlas, Strategy architect. 8.67 tok/s, 6.68 GB at 8,192 ctx. Configuration-derived from Qwen3-8B; not THOX-trained. Apache-2.0.
tools thinking11 Pulls 1 Tag Updated 2 days ago
-
thox-ios-mesh-8b
iOS mesh-assisted node model (never a silent fallback). THOX-configured Qwen3-8B, not THOX-trained. 8.38 tok/s long-form, 6.68 GB RAM on CPU at 8,192 ctx; grounding 7/8, trap declined. Apache-2.0.
tools thinking11 Pulls 1 Tag Updated yesterday
-
thoxmeshgroup-32b
Workgroup assistant on dense Qwen3-32B with THOX profiles and Knowledge Pack. 5.36 tok/s long-form, 22.61 GB footprint on a GPU host at 8,192 ctx; grounding 8/8, trap declined. Config-only, not THOX-trained. Apache-2.0.
tools thinking11 Pulls 1 Tag Updated yesterday
-
thoxclip-9m-role
ThoxClip Device-Role Model — the device-role model for the ThoxClip persona, fine-tuned from a THOX model trained from scratch (no external base) on ThoxClip's own role corpus GGUF: TQ1_0
11 Pulls 1 Tag Updated 2 months ago
-
thoxmesh-moe-30b
MeshStack routing assistant on sparse Qwen3-30B-A3B (MoE) with THOX Knowledge Pack. 53.75 tok/s long-form, 14.99 GB footprint on a GPU host at 8,192 ctx; grounding 8/8, trap declined. Config-only, not THOX-trained. Apache-2.0.
tools thinking11 Pulls 1 Tag Updated yesterday
-
thoxmesh-open-14b
MeshStack routing assistant on Qwen3-14B with the THOX Knowledge Pack. 25.25 tok/s long-form, 10.49 GB footprint on a GPU host at 8,192 ctx; grounding 8/8, trap declined. Config-only, not THOX-trained. Apache-2.0.
tools thinking11 Pulls 1 Tag Updated yesterday
-
thoxagent-route-0.6b
THOX-configured Qwen3-0.6B agent model for classification and structured route proposal, with the THOX agent system profile and Knowledge Pack v0.1. Measured 64.23 tok/s, 1.53 GB footprint (CPU, 2026-10-31). Not THOX-trained. Apache-2.0.
tools thinking11 Pulls 1 Tag Updated 2 days ago
-
thoxhuman-reason-14b
Persona-following reasoning model on Qwen3-14B with THOX profiles and Knowledge Pack. 27.67 tok/s long-form, 10.49 GB footprint on a GPU host at 8,192 ctx; grounding 8/8, trap declined. Config-only, not THOX-trained. Apache-2.0.
tools thinking10 Pulls 1 Tag Updated yesterday
-
thoxopen-8b
THOX Open laptop model: Qwen3-8B packaged by THOX with the THOX Open system profile and THOX Knowledge Pack v0.1. Configuration-derived; THOX did not train the weights. Apache-2.0.
tools thinking10 Pulls 1 Tag Updated 2 days ago
-
thoxair-chat-0.5b
ThoxAir chat configuration for the ThoxAir (2 GB) target (Qwen2.5-0.5B, THOX-Configured) — thoxAir chat configuration for the ThoxAir (2 GB) target 58.8 tok/s long-form (measured 2026-10-31, conditions under Evaluation) · Ollama Modelfile
tools10 Pulls 1 Tag Updated 2 days ago
-
thox-cluster-code
A THOX configuration and distribution of Alibaba Cloud's Qwen2.5-Coder 32B, published for local software-engineering work and MeshStack deployments. Apache-2.0.
10 Pulls 1 Tag Updated 10 months ago
-
thox-ava
THOX digital-human persona: Ava, Primary DigitalHuman, platforms and fleet orchestration. 8.99 tok/s, 6.68 GB at 8,192 ctx. Configuration-derived from Qwen3-8B; not THOX-trained. Apache-2.0.
tools thinking10 Pulls 1 Tag Updated 2 days ago
-
thox-ginger
THOX digital-human persona: Ginger Weaving, Chief Innovation & Engineering Officer. 13.75 tok/s, 3.89 GB at 8,192 ctx. Configuration-derived from Qwen3-4B; not THOX-trained. Apache-2.0.
tools thinking10 Pulls 1 Tag Updated 2 days ago
-
thoxagent-edge-1.7b
THOX-configured Qwen3-1.7B agent model for local tool-plan drafting, with the THOX agent system profile and Knowledge Pack v0.1. Measured 14.52 tok/s, 2.43 GB footprint (CPU, 2026-10-31). Not THOX-trained. Apache-2.0.
tools thinking10 Pulls 1 Tag Updated 2 days ago
-
thox-thoxy
THOX digital-human persona: THOXY, THOX platform guide. 8.34 tok/s, 3.89 GB at 8,192 ctx. Configuration-derived from Qwen3-4B; not THOX-trained. Apache-2.0.
tools thinking10 Pulls 1 Tag Updated 2 days ago
-
thoxmeshgroup-30b-a3b
Workgroup assistant on sparse Qwen3-30B-A3B (MoE) with THOX profiles and Knowledge Pack. 53.68 tok/s long-form, 14.99 GB footprint on a GPU host at 8,192 ctx; grounding 8/8, trap declined. Config-only, not THOX-trained. Apache-2.0.
tools thinking9 Pulls 1 Tag Updated yesterday
-
thox-cluster-secure
A THOX configuration of Qwen2.5 72B with a security-focused system prompt, published for local inference in hardened, network-isolated environments. No certification or authorization is claimed. Qwen Research License (non-commercial).
9 Pulls 1 Tag Updated 10 months ago
-
thox-pro-8b
A THOX configuration and distribution of Mistral AI's Ministral 3 8B: a multimodal (text and image) model that runs locally on a single device. Apache-2.0.
vision tools9 Pulls 1 Tag Updated 10 months ago
-
thox-ios-chat-3b
iOS grounded chat model. THOX-configured Qwen2.5-3B, not THOX-trained. 11.32 tok/s long-form, 2.42 GB RAM on CPU at 8,192 ctx; grounding 6/8, trap declined. Qwen Research License (non-commercial).
tools9 Pulls 1 Tag Updated 2 days ago
-
thox-pro-14b
A THOX configuration and distribution of Mistral AI's Ministral 3 14B: a multimodal (text and image) model for professional document and image work, running locally on a single device. Apache-2.0.
vision tools9 Pulls 1 Tag Updated 10 months ago
-
thox-cluster-maverick
A THOX configuration and distribution of Alibaba Cloud's Qwen3 235B MoE, the largest model in the MagStack cluster family, published for local inference and MeshStack deployments. Apache-2.0.
9 Pulls 1 Tag Updated 10 months ago
-
thoxkey-9m-role
ThoxKey Device Model — the device-role model for ThoxKey, fine-tuned from a THOX model trained from scratch (no external base) Safetensors weights · GGUF: Q8_0, TQ1_0, TQ2_0
9m9 Pulls 1 Tag Updated 3 days ago
-
thox-rhea
THOX digital-human persona: Rhea, People manager. 8.85 tok/s, 6.68 GB at 8,192 ctx. Configuration-derived from Qwen3-8B; not THOX-trained. Apache-2.0.
tools thinking9 Pulls 1 Tag Updated 2 days ago
-
thox-mira
THOX digital-human persona: Mira, Knowledge SME. 13.87 tok/s, 3.89 GB at 8,192 ctx. Configuration-derived from Qwen3-4B; not THOX-trained. Apache-2.0.
tools thinking9 Pulls 1 Tag Updated 2 days ago
-
thox-kai
THOX digital-human persona: Kai, Device worker. 13.78 tok/s, 3.89 GB at 8,192 ctx. Configuration-derived from Qwen3-4B; not THOX-trained. Apache-2.0.
tools thinking9 Pulls 1 Tag Updated 2 days ago
-
thox-ios-system
THOX iOS system profile over Qwen3-1.7B (stand-in: Apple Foundation Models weights are not distributed). THOX-configured, not THOX-trained. 24.37 tok/s long-form, 2.43 GB RAM on CPU at 8,192 ctx; grounding 6/8, trap declined. Apache-2.0.
tools thinking9 Pulls 1 Tag Updated 2 days ago
-
thoxair-route-0.5b
ThoxAir routing configuration for the ThoxAir (2 GB) target (Qwen2.5-0.5B, THOX-Configured) — thoxAir routing configuration for the ThoxAir (2 GB) target 55.35 tok/s long-form (measured 2026-10-31, conditions under Evaluation) · Ollama Modelfile
tools9 Pulls 1 Tag Updated 2 days ago
-
thox-cluster-swift
A THOX configuration and distribution of Mistral AI's Ministral 3 8B, the smallest and lightest model in the MagStack cluster family, published for local inference and MeshStack deployments. Apache-2.0.
vision tools9 Pulls 1 Tag Updated 10 months ago
-
thox-cluster-deep
A THOX configuration and distribution of Meta's Llama 3.1 405B, published for local research-scale inference and MeshStack deployments. Llama 3.1 Community License.
9 Pulls 1 Tag Updated 10 months ago
-
thox-nova
THOX digital-human persona: Nova, Coding and build. 8.91 tok/s, 6.68 GB at 8,192 ctx. Configuration-derived from Qwen3-8B; not THOX-trained. Apache-2.0.
tools thinking9 Pulls 1 Tag Updated 2 days ago
-
thox-ios-micro-0.5b
iOS micro model for intent classification. THOX-configured Qwen2.5-0.5B, not THOX-trained. 61.97 tok/s long-form, 0.59 GB RAM on CPU at 8,192 ctx; grounding 5/8, trap declined. Apache-2.0.
tools9 Pulls 1 Tag Updated 2 days ago
-
thoxmesh-agent-30b
Agent-profile model on sparse Qwen3-30B-A3B (MoE) with THOX profiles and Knowledge Pack. 52.23 tok/s long-form, 14.99 GB footprint on a GPU host at 8,192 ctx; grounding 8/8, trap declined. Config-only, not THOX-trained. Apache-2.0.
tools thinking9 Pulls 1 Tag Updated yesterday
-
thoxmini-mind-1.5b
ThoxMini decision configuration (4 GB target). THOX-configured Qwen2.5-1.5B, not THOX-trained. 11.35 tok/s long-form, 1.37 GB RAM on CPU at 8,192 ctx; grounding 7/8, trap declined. Apache-2.0.
tools9 Pulls 1 Tag Updated 2 days ago
-
thox-sera
THOX digital-human persona: Sera, Compliance judge. 8.11 tok/s, 6.68 GB at 8,192 ctx. Configuration-derived from Qwen3-8B; not THOX-trained. Apache-2.0.
tools thinking9 Pulls 1 Tag Updated 2 days ago
-
thox-ios-edge-1.5b
iOS edge assistant and routing model. THOX-configured Qwen2.5-1.5B, not THOX-trained. 35.67 tok/s long-form, 1.37 GB RAM on CPU at 8,192 ctx; grounding 7/8, trap declined. Apache-2.0.
tools8 Pulls 1 Tag Updated 2 days ago
-
thox-pro-32b
A THOX configuration and distribution of Alibaba Cloud's Qwen3 32B, the largest single-device model in the Pro family, published for local professional work. Apache-2.0.
8 Pulls 1 Tag Updated 10 months ago
-
thox-ios-open-4b
iOS open chat configuration (profile-only, not a fine-tune). THOX-configured Qwen3-4B, not THOX-trained. 15.34 tok/s long-form, 3.89 GB RAM on CPU at 8,192 ctx; grounding 8/8, trap declined. Apache-2.0.
tools thinking8 Pulls 1 Tag Updated yesterday
-
thoxwork-office-4b
THOX Work office and document assistant. THOX-configured Qwen3-4B, not THOX-trained. 15.32 tok/s long-form, 3.89 GB RAM on CPU at 8,192 ctx; grounding 7/8, trap declined. Apache-2.0.
tools thinking8 Pulls 1 Tag Updated yesterday
-
thoxedge-4b
THOX Edge laptop configuration (not the trained ThoxEdge-0.8B). THOX-configured Qwen3-4B, not THOX-trained. 15.19 tok/s long-form, 3.89 GB RAM on CPU at 8,192 ctx; grounding 8/8, trap declined. Apache-2.0.
tools thinking8 Pulls 1 Tag Updated yesterday
-
thoxmini-chat-3b
ThoxMini chat configuration (4 GB target). THOX-configured Qwen2.5-3B, not THOX-trained. 11.66 tok/s long-form, 2.42 GB RAM on CPU at 8,192 ctx; grounding 6/8, trap declined. Qwen Research License (non-commercial).
tools8 Pulls 1 Tag Updated 2 days ago
-
thoxwork-analyst-8b
THOX Work analyst model. THOX-configured Qwen3-8B, not THOX-trained. 8.73 tok/s long-form, 6.68 GB RAM on CPU at 8,192 ctx; grounding 8/8, trap declined. Apache-2.0.
tools thinking8 Pulls 1 Tag Updated yesterday
-
thox-cluster-scout
A THOX configuration and distribution of Alibaba Cloud's Qwen2.5 72B, published for local inference and MeshStack deployments at department scale. Qwen Research License (non-commercial).
8 Pulls 1 Tag Updated 10 months ago
-
thoxcode-8b
THOX Code model: Qwen2.5-Coder-7B packaged by THOX with the THOX Code system profile and THOX Knowledge Pack v0.1. Measured 9.2 tok/s, 5.5 GB RAM. Configuration-derived; THOX did not train the weights. Apache-2.0.
tools8 Pulls 1 Tag Updated 2 days ago
-
thoxclip-chat-3b
ThoxClip (6 GB) target chat configuration (Qwen2.5-3B, THOX-Configured) — thoxClip (6 GB) target chat configuration 11.03 tok/s long-form (measured 2026-10-31, conditions under Evaluation) · Ollama Modelfile · Live on Ollama
tools8 Pulls 1 Tag Updated 2 days ago
-
thox-helix
THOX digital-human persona: Helix, Finance SME. 9.14 tok/s, 6.68 GB at 8,192 ctx. Configuration-derived from Qwen3-8B; not THOX-trained. Apache-2.0.
tools thinking8 Pulls 1 Tag Updated 2 days ago
-
thoxwork-mind-4b
THOX Work decision-profile model. THOX-configured Qwen3-4B, not THOX-trained. 15.32 tok/s long-form, 3.89 GB RAM on CPU at 8,192 ctx; grounding 8/8, trap declined. Apache-2.0.
tools thinking7 Pulls 1 Tag Updated yesterday
-
thoxclip-agent-4b
ThoxClip agent configuration (6 GB target). THOX-configured Qwen3-4B, not THOX-trained. 15.28 tok/s long-form, 3.89 GB RAM on CPU at 8,192 ctx; grounding 8/8, trap declined. Apache-2.0.
tools thinking7 Pulls 1 Tag Updated yesterday
-
thoxagent-core-4b
THOX agent core model. THOX-configured Qwen3-4B, not THOX-trained. 15.12 tok/s long-form, 3.89 GB RAM on CPU at 8,192 ctx; grounding 8/8, trap declined. Apache-2.0.
tools thinking6 Pulls 1 Tag Updated yesterday
-
thoxwork-agent-8b
THOX Work agent model. THOX-configured Qwen3-8B, not THOX-trained. 8.59 tok/s long-form, 6.68 GB RAM on CPU at 8,192 ctx; grounding 8/8, trap declined. Apache-2.0.
tools thinking6 Pulls 1 Tag Updated yesterday
-
thoxclip-mind-1.5b
ThoxClip (6 GB) target configuration (Qwen2.5-1.5B, THOX-Configured) — thoxClip (6 GB) target configuration: on-device decision support 20.87 tok/s long-form (measured 2026-10-31, conditions under Evaluation) · Ollama Modelfile · Live on Ollama
tools6 Pulls 1 Tag Updated 2 days ago
-
thoxopen-4b
THOX Open small laptop model: Qwen3-4B packaged by THOX with the THOX Open system profile and THOX Knowledge Pack v0.1. Measured 10.7 tok/s long-form, 3.9 GB RAM. Configuration-derived; THOX did not train the weights. Apache-2.0.
tools thinking6 Pulls 1 Tag Updated 2 days ago
-
thoxmeshgroup-federated-8b
MeshGroup federated routing profile (request routing only). THOX-configured Qwen3-8B, not THOX-trained. 8.75 tok/s long-form, 6.68 GB RAM on CPU at 8,192 ctx; grounding 8/8, trap declined. Apache-2.0.
tools thinking5 Pulls 1 Tag Updated yesterday