🔍 Model Trust Scorecard

Transparent, reproducible verification of AI model benchmark claims

Models Evaluated

28.8

Average Trust Score

Total Claims

15%

Claims Verified

Rankings

Rank	Model (Vendor)	Parameters / Context	Trust Score (Verified)	Capabilities	Use-Case Strengths	License
1	Claude Opus 4.5 Anthropic	- 200K ctx	33.2 1/7 verified	Code	coding: 89.3, reasoning: 88.3, safety: 69.8	unknown
2	Llama 3.1 405B Meta	405.0B 128K ctx	51.8 1/5 verified	Code • Tools • Multi-Lang	coding: 89.0, reasoning: 86.5, multilingual: 87.5	unknown
3	DeepSeek V3.2 DeepSeek	671.0B 128K ctx	57.1 2/6 verified	Code	coding: 80.8, reasoning: 84.9	unknown
4	MiniMax M2.5 Cloud MiniMax	32.0B 128K ctx	63.8 1/1 verified	Code • Agent	coding: 80.2	unknown
5	Kimi K2.5 Cloud Moonshot	32.0B 256K ctx	9.8 0/1 verified	Vision • Code • Agent	coding: 79.8	unknown
6	GLM 5.1 Cloud Z.ai	32.0B 128K ctx	9.8 0/1 verified	Code • Agent	coding: 78.5	unknown
7	MiniMax M2.7 Cloud MiniMax	32.0B 128K ctx	9.8 0/1 verified	Code	coding: 78.0	unknown
8	Gemini 2.5 Pro Google	- 2097K ctx	29.8 0/6 verified	Code	coding: 75.0, reasoning: 84.8, safety: 68.1	unknown
9	GPT-4.1 OpenAI	- 128K ctx	43.7 2/5 verified	Code	coding: 72.8, reasoning: 86.5	unknown
10	Devstral 2 123B Cloud Mistral	123.0B 128K ctx	9.8 0/1 verified	Code • Tools • Agent	coding: 72.2	unknown
11	DeepSeek R1 14B DeepSeek	14.0B 128K ctx	22.2 0/4 verified	Code	coding: 57.6, reasoning: 81.2	unknown
12	Qwen3 14B Alibaba	14.0B 128K ctx	26.0 0/5 verified	Code • Tools • Agent • Multi-Lang	reasoning: 79.2, multilingual: 81.1	unknown
13	Llama 3.2 Vision 11B Meta	11.0B 128K ctx	22.2 0/3 verified	Vision	reasoning: 52.6	unknown
14	Hermes 3 8B Nous	8.0B 128K ctx	14.8 0/1 verified	Code • Tools • Agent	commonsense: 3.1	unknown
15	GLM-4.7-Flash Zhipu AI	9.0B 128K ctx	N/A 0/0 verified	Multi-Lang	-	unknown
16	GLM-4-32B-0414-128K Zhipu AI	32.0B 128K ctx	N/A 0/0 verified	Multi-Lang	-	unknown
17	GLM-4.6V-Flash Zhipu AI	9.0B 128K ctx	N/A 0/0 verified	Vision • Multi-Lang	-	unknown
18	GLM-4.5-Air Zhipu AI	9.0B 128K ctx	N/A 0/0 verified	Multi-Lang	-	unknown
19	Nomic Embed Text Nomic	0.137B 8K ctx	N/A 0/0 verified	-	-	unknown
20	Cogito 2.1 671B Cloud Deep Cogito	671.0B 128K ctx	N/A 0/0 verified	Code	-	unknown
21	AutoGLM-Phone-Multilingual Zhipu AI	9.0B 32K ctx	N/A 0/0 verified	Tools • Agent • Multi-Lang	-	unknown
22	GLM-OCR Zhipu AI	9.0B 32K ctx	N/A 0/0 verified	Vision • Multi-Lang	-	unknown
23	GLM-4.5-AirX Zhipu AI	9.0B 128K ctx	N/A 0/0 verified	Multi-Lang	-	unknown
24	Granite 3.3 8B IBM	8.0B 128K ctx	N/A 0/0 verified	Code	-	unknown
25	GLM-4.7-FlashX Zhipu AI	9.0B 128K ctx	N/A 0/0 verified	Multi-Lang	-	unknown
26	Cogito 14B Deep Cogito	14.0B 128K ctx	N/A 0/0 verified	Code	-	unknown
27	CogVideoX-3 Zhipu AI	- 8K ctx	N/A 0/0 verified	-	-	unknown
28	LLaVA Community	7.0B 4K ctx	N/A 0/0 verified	Vision	-	unknown
29	GLM-4.6V Zhipu AI	32.0B 128K ctx	N/A 0/0 verified	Vision • Multi-Lang	-	unknown
30	GLM-4.6V-FlashX Zhipu AI	9.0B 128K ctx	N/A 0/0 verified	Vision • Multi-Lang	-	unknown
31	GLM-5-Turbo Zhipu AI	9.0B 128K ctx	N/A 0/0 verified	Multi-Lang	-	unknown
32	GLM-ASR-2512 Zhipu AI	- 8K ctx	N/A 0/0 verified	-	-	unknown
33	GLM-4.5-Flash Zhipu AI	9.0B 128K ctx	N/A 0/0 verified	Multi-Lang	-	unknown
34	Qwen3 VL 235B Cloud Alibaba	235.0B 128K ctx	N/A 0/0 verified	Vision • Multi-Lang	-	unknown
35	GLM-4.5 Zhipu AI	9.0B 128K ctx	N/A 0/0 verified	Multi-Lang	-	unknown
36	GLM-4.6 Zhipu AI	32.0B 128K ctx	N/A 0/0 verified	Multi-Lang	-	unknown
37	Gemma 3 4B Google	4.0B 128K ctx	N/A 0/0 verified	Vision • Multi-Lang	-	unknown
38	Nomic Embed Text v2 MoE Nomic	0.55B 8K ctx	N/A 0/0 verified	Multi-Lang	-	unknown
39	GLM-5V-Turbo Zhipu AI	9.0B 128K ctx	N/A 0/0 verified	Vision • Multi-Lang	-	unknown
40	GLM-4.5V Zhipu AI	9.0B 128K ctx	N/A 0/0 verified	Vision • Multi-Lang	-	unknown
41	MXBAI Embed Large Mixedbread	0.335B 0K ctx	N/A 0/0 verified	-	-	unknown
42	Embedding Gemma Google	0.3B 8K ctx	N/A 0/0 verified	-	-	unknown
43	DeepCoder Community	14.0B 8K ctx	N/A 0/0 verified	Code	-	unknown
44	Qwen3 Embedding 0.6B Alibaba	0.6B 32K ctx	N/A 0/0 verified	Multi-Lang	-	unknown
45	GLM-4.7 Zhipu AI	32.0B 128K ctx	N/A 0/0 verified	Multi-Lang	-	unknown
46	GLM-5 Zhipu AI	9.0B 128K ctx	N/A 0/0 verified	Multi-Lang	-	unknown
47	GLM-5.1 Zhipu AI	32.0B 256K ctx	N/A 0/0 verified	Multi-Lang	-	unknown
48	CogView-4-250304 Zhipu AI	- 8K ctx	N/A 0/0 verified	Vision	-	unknown
49	GLM-Image Zhipu AI	- 8K ctx	N/A 0/0 verified	Vision	-	unknown
50	GLM-4-Plus Zhipu AI	9.0B 128K ctx	N/A 0/0 verified	Multi-Lang	-	unknown

Where results go

This page is generated into docs/index.html for GitHub Pages.
The machine-readable aggregate is committed as trust_scores.json.
A markdown summary is committed as trust_scores.md.
Per-model verification reports are attached to the workflow run as artifacts.

Request a review for a model not listed here

Open the Model Submission issue.
Or submit a PR that adds models/<model-id>.json to the catalog.
For one-off checks, run the CLI with pasted claims via trust-scorecard score --text or --text-file.
Raw ollama list inventories and categorized Markdown lists are valid input examples. Models that do not map to a catalog entry are skipped until catalog data or pasted claims are provided.