OpenAI compatible API · Attested · Public status

Compare AI models

Compare context, provider routes, pricing, privacy posture, and measured performance across the TrustedRouter model catalog.

Verify gateway

Onebase URL to migrate

100sof models and routes

0prompt or output logs. Always.

Measured speed

Find the fastest routes.

Compare provider TTFT, throughput, and uptime from routed probes.

Open leaderboard → Open weights

Compare open models.

Browse DeepSeek, GLM, Gemma, Kimi, MiniMax, Qwen, and other open routes.

Browse models → Privacy

Require ZDR or confidential routes.

Separate gateway attestation from each downstream provider's privacy posture.

Review privacy →

Current comparison, preserved evidence.

Comparison pages use the current catalog and live route measurements. Monthly reports preserve the historical evidence after prices and provider performance change.

Monthly reports Run your own eval

Model comparisons

Pick two models. Compare the routes that actually serve them.

Every comparison uses the current TrustedRouter catalog, prices, provider posture, and measured route data.

Comparison	First model	Routes	Second model	Routes
DeepSeek: DeepSeek V4.1 Flash vs Ministral 3 14B Instruct	`deepseek/deepseek-v4.1-flash`	37	`mistralai/ministral-3-14b-instruct`	2
Mistral: Ministral 3 8B 2512 vs OpenAI: gpt-oss-120b	`mistralai/ministral-8b-2512`	2	`openai/gpt-oss-120b`	37
Mistral: Mistral Large 3 2512 vs Qwen: Qwen3.8 27B	`mistralai/mistral-large-2512`	1	`qwen/qwen3.8-27b`	30
Mistral: Mistral Medium 3.5 vs MoonshotAI: Kimi K2.6	`mistralai/mistral-medium-3-5`	2	`moonshotai/kimi-k2.6`	27
Google: Gemma 4 31B vs MoonshotAI: Kimi K2 0905	`google/gemma-4-31b-it`	26	`moonshotai/kimi-k2-0905`	2
DeepSeek: DeepSeek V4 Pro 0423 vs moonshotai/Kimi-K2-Instruct-0905	`deepseek/deepseek-v4-pro`	26	`moonshotai/kimi-k2-instruct-0905`	2
morph-v3-fast vs Z.ai: GLM 5.3	`morph/morph-v3-fast`	1	`z-ai/glm-5.3`	50
morph-v3-large vs Z.ai: GLM 5.2	`morph/morph-v3-large`	1	`z-ai/glm-5.2`	48
nvidia/Nemotron-3-Nano-Omni-Reasoning-30B-A3B vs Z.ai: GLM 5.3 Flash	`nvidia/nemotron-3-nano-omni-reasoning-30b-a3b`	2	`z-ai/glm-5.3-flash`	48
MoonshotAI: Kimi K3 vs Nemotron 3 Super	`moonshotai/kimi-k3`	45	`nvidia/nemotron-3-super`	1
DeepSeek: DeepSeek V4 Flash 0731 vs Ternary Bonsai 27B	`deepseek/deepseek-v4-flash-0731`	35	`prism-ml/ternary-bonsai-27b`	2
DeepSeek: DeepSeek V4.1 Flash vs Qwen 3.8 2.4T	`deepseek/deepseek-v4.1-flash`	37	`qwen/qwen-3-8-2-4t-a95b`	2
OpenAI: gpt-oss-120b vs Qwen VL OCR	`openai/gpt-oss-120b`	37	`qwen/qwen-vl-ocr`	1
Qwen VL OCR 2025 11 20 vs Qwen: Qwen3.8 27B	`qwen/qwen-vl-ocr-2025-11-20`	1	`qwen/qwen3.8-27b`	30
MoonshotAI: Kimi K2.6 vs Qwen: Qwen3 235B A22B Instruct 2507	`moonshotai/kimi-k2.6`	27	`qwen/qwen3-235b-a22b-2507`	1
Google: Gemma 4 31B vs Qwen3-4B-Instruct-2507	`google/gemma-4-31b-it`	26	`qwen/qwen3-4b-instruct-2507`	1
DeepSeek: DeepSeek V4 Pro 0423 vs Qwen: Qwen3 VL 8B Instruct	`deepseek/deepseek-v4-pro`	26	`qwen/qwen3-vl-8b-instruct`	2
MiniMax: MiniMax M3 vs Qwen3 VL Plus	`minimax/minimax-m3`	25	`qwen/qwen3-vl-plus`	1
Qwen3 VL Plus 2025 09 23 vs Z.ai: GLM 5.3	`qwen/qwen3-vl-plus-2025-09-23`	1	`z-ai/glm-5.3`	50
Qwen3 VL Plus 2025 12 19 vs Z.ai: GLM 5.2	`qwen/qwen3-vl-plus-2025-12-19`	1	`z-ai/glm-5.2`	48
MoonshotAI: Kimi K3 vs Qwen3.6 Plus 2026-04-02	`moonshotai/kimi-k3`	45	`qwen/qwen3.6-plus-2026-04-02`	2
DeepSeek: DeepSeek V4 Flash 0731 vs Union Alpha	`deepseek/deepseek-v4-flash-0731`	35	`stealth/union-alpha`	1
DeepSeek: DeepSeek V4.1 Flash vs StepFun: Step 3.5 Flash	`deepseek/deepseek-v4.1-flash`	37	`stepfun/step-3.5-flash`	1
OpenAI: gpt-oss-120b vs Step 3.5 Flash 2603	`openai/gpt-oss-120b`	37	`stepfun/step-3.5-flash-2603`	1
Qwen: Qwen3.8 27B vs 0GM-1.0-35B-A3B	`qwen/qwen3.8-27b`	30	`zero-g/0gm-1.0-35b-a3b`	1
Inception: Mercury 2.5 vs MoonshotAI: Kimi K2.6	`inception/mercury-2.5`	1	`moonshotai/kimi-k2.6`	27
ByteDance/Seed-1.8 vs Google: Gemma 4 31B	`bytedance/seed-1.8`	2	`google/gemma-4-31b-it`	26
ByteDance/Seed-2.0-code vs DeepSeek: DeepSeek V4 Pro 0423	`bytedance/seed-2.0-code`	2	`deepseek/deepseek-v4-pro`	26
ByteDance/Seed-2.0-mini vs MiniMax: MiniMax M3	`bytedance/seed-2.0-mini`	2	`minimax/minimax-m3`	25
ByteDance/Seed-2.0-pro vs Z.ai: GLM 5.3	`bytedance/seed-2.0-pro`	2	`z-ai/glm-5.3`	50
Gemma 4 Uncensored vs Z.ai: GLM 5.2	`google/gemma-4-uncensored`	2	`z-ai/glm-5.2`	48
Mistral: Codestral 2508 vs Z.ai: GLM 5.3 Flash	`mistralai/codestral-2508`	2	`z-ai/glm-5.3-flash`	48
Kimi K2.5 vs MoonshotAI: Kimi K3	`moonshotai/kimi-k2-5`	2	`moonshotai/kimi-k3`	45
DeepSeek: DeepSeek V4 Flash 0731 vs Kimi K2.6	`deepseek/deepseek-v4-flash-0731`	35	`moonshotai/kimi-k2-6`	2
DeepSeek: DeepSeek V4.1 Flash vs Kimi K2.7 Code	`deepseek/deepseek-v4.1-flash`	37	`moonshotai/kimi-k2-7-code`	2
@cf/nvidia/nemotron-3-120b-a12b vs OpenAI: gpt-oss-120b	`nvidia/nemotron-3-120b-a12b`	1	`openai/gpt-oss-120b`	37
NVIDIA-Nemotron-3-Nano-30B-A3B-BF16 vs Qwen: Qwen3.8 27B	`nvidia/nemotron-3-nano-30b-a3b-bf16`	1	`qwen/qwen3.8-27b`	30
MoonshotAI: Kimi K2.6 vs Qwen 3.5 35B A3B	`moonshotai/kimi-k2.6`	27	`qwen/qwen3-5-35b-a3b`	2
Google: Gemma 4 31B vs Qwen 3.6 35B A3B	`google/gemma-4-31b-it`	26	`qwen/qwen3-6-35b-a3b`	2
DeepSeek: DeepSeek V4 Pro 0423 vs Qwen 3 Next 80b	`deepseek/deepseek-v4-pro`	26	`qwen/qwen3-next-80b`	2
MiniMax: MiniMax M3 vs StepFun: Step-3.7-Flash	`minimax/minimax-m3`	25	`stepfun-ai/step-3.7-flash`	2
Z.ai: GLM 5.3 vs GLM 4.6	`z-ai/glm-5.3`	50	`zai-org/glm-4.6`	2
Z.ai: GLM 5.2 vs GLM 4.7	`z-ai/glm-5.2`	48	`zai-org/glm-4.7`	2
Z.ai: GLM 5.3 Flash vs GLM 5.1	`z-ai/glm-5.3-flash`	48	`zai-org/GLM-5.1`	2
MoonshotAI: Kimi K3 vs GLM 5.1	`moonshotai/kimi-k3`	45	`zai-org/glm-5.1`	2
DeepSeek: DeepSeek V4 Flash 0731 vs GLM 5	`deepseek/deepseek-v4-flash-0731`	35	`zai-org/glm-5`	2
DeepSeek: DeepSeek V4.1 Flash vs Z.ai: GLM 5 Turbo	`deepseek/deepseek-v4.1-flash`	37	`z-ai/glm-5-turbo`	3
Anthropic: Claude Haiku 4.5 vs OpenAI: gpt-oss-120b	`anthropic/claude-haiku-4.5`	2	`openai/gpt-oss-120b`	37
Anthropic: Claude Opus 4.5 vs Qwen: Qwen3.8 27B	`anthropic/claude-opus-4.5`	2	`qwen/qwen3.8-27b`	30
MoonshotAI: Kimi K2.6 vs OpenAI: o1	`moonshotai/kimi-k2.6`	27	`openai/o1`	2
Google: Gemma 4 31B vs OpenAI: o3	`google/gemma-4-31b-it`	26	`openai/o3`	2
DeepSeek: DeepSeek V4 Pro 0423 vs OpenAI: o3 Mini	`deepseek/deepseek-v4-pro`	26	`openai/o3-mini`	2
MiniMax: MiniMax M3 vs OpenAI: o4 Mini	`minimax/minimax-m3`	25	`openai/o4-mini`	2
Z.ai: GLM 5.3 vs GLM 4.7 Flash	`z-ai/glm-5.3`	50	`zai-org/glm-4.7-flash`	2
MiniMax M2.5 vs Z.ai: GLM 5.2	`minimax/minimax-m25`	2	`z-ai/glm-5.2`	48
MiniMax M2.7 vs Z.ai: GLM 5.3 Flash	`minimax/minimax-m27`	2	`z-ai/glm-5.3-flash`	48
DeepSeek: DeepSeek V3.1 Terminus vs MoonshotAI: Kimi K3	`deepseek/deepseek-v3.1-terminus`	1	`moonshotai/kimi-k3`	45
DeepSeek: DeepSeek V3.2 Exp vs DeepSeek: DeepSeek V4 Flash 0731	`deepseek/deepseek-v3.2-exp`	1	`deepseek/deepseek-v4-flash-0731`	35
DeepSeek: DeepSeek V4.1 Flash vs Meta: Llama Guard 4 12B	`deepseek/deepseek-v4.1-flash`	37	`meta-llama/llama-guard-4-12b`	2
AionLabs: Aion 2.0 vs OpenAI: gpt-oss-120b	`aion-labs/aion-2.0`	1	`openai/gpt-oss-120b`	37
AionLabs: Aion 3.0 vs Qwen: Qwen3.8 27B	`aion-labs/aion-3.0`	1	`qwen/qwen3.8-27b`	30
AionLabs: Aion 3.0 Mini vs MoonshotAI: Kimi K2.6	`aion-labs/aion-3.0-mini`	1	`moonshotai/kimi-k2.6`	27
Baichuan M2 32B vs Google: Gemma 4 31B	`baichuan/baichuan-m2-32b`	2	`google/gemma-4-31b-it`	26
CoBuddy vs DeepSeek: DeepSeek V4 Pro 0423	`baidu/cobuddy`	2	`deepseek/deepseek-v4-pro`	26
Doubao Seed Character vs MiniMax: MiniMax M3	`bytedance/doubao-seed-character-260628`	1	`minimax/minimax-m3`	25
OpenAI GPT OSS vs Z.ai: GLM 5.3	`cerebras/gpt-oss-120b`	2	`z-ai/glm-5.3`	50
DeepSeek-R1-Distill-Qwen-14B vs Z.ai: GLM 5.2	`deepseek/deepseek-r1-distill-qwen-14b`	1	`z-ai/glm-5.2`	48
Google: Gemma 3 4B vs Z.ai: GLM 5.3 Flash	`google/gemma-3-4b-it`	2	`z-ai/glm-5.3-flash`	48
google/gemma-4-31B-it-Ultra vs MoonshotAI: Kimi K3	`google/gemma-4-31b-it-ultra`	2	`moonshotai/kimi-k3`	45
DeepSeek: DeepSeek V4 Flash 0731 vs google/gemma-4-31B-turbo	`deepseek/deepseek-v4-flash-0731`	35	`google/gemma-4-31b-turbo`	2
DeepSeek: DeepSeek V4.1 Flash vs inclusionAI: Ling 3.0 Flash VL	`deepseek/deepseek-v4.1-flash`	37	`inclusionai/ling-3.0-flash-vl`	2
meta-llama/Llama-3.3-70B-Instruct-Turbo vs OpenAI: gpt-oss-120b	`meta-llama/llama-3.3-70b-instruct-turbo`	2	`openai/gpt-oss-120b`	37
meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo vs Qwen: Qwen3.8 27B	`meta-llama/meta-llama-3.1-8b-instruct-turbo`	2	`qwen/qwen3.8-27b`	30
Mistral: Ministral 3 3B 2512 vs MoonshotAI: Kimi K2.6	`mistralai/ministral-3b-2512`	2	`moonshotai/kimi-k2.6`	27
Google: Gemma 4 31B vs Kimi K2 Instruct	`google/gemma-4-31b-it`	26	`moonshotai/kimi-k2-instruct`	2
DeepSeek: DeepSeek V4 Pro 0423 vs Nous: Hermes 3 405B Instruct	`deepseek/deepseek-v4-pro`	26	`nousresearch/hermes-3-llama-3.1-405b`	2
MiniMax: MiniMax M3 vs Hermes 4 405B	`minimax/minimax-m3`	25	`NousResearch/Hermes-4-405B`	2
Nemotron 3 Ultra vs Z.ai: GLM 5.3	`nvidia/nemotron-3-ultra-550b`	2	`z-ai/glm-5.3`	50
nvidia/Nemotron-Content-Safety-3.5 vs Z.ai: GLM 5.2	`nvidia/nemotron-content-safety-3.5`	2	`z-ai/glm-5.2`	48
openai/gpt-oss-120b-Ultra vs Z.ai: GLM 5.3 Flash	`openai/gpt-oss-120b-ultra`	2	`z-ai/glm-5.3-flash`	48
MoonshotAI: Kimi K3 vs GPT-OSS Safeguard 120B	`moonshotai/kimi-k3`	45	`openai/gpt-oss-safeguard-120b`	2
DeepSeek: DeepSeek V4 Flash 0731 vs Qwen MT Flash	`deepseek/deepseek-v4-flash-0731`	35	`qwen/qwen-mt-flash`	1
DeepSeek: DeepSeek V4.1 Flash vs Qwen MT Lite	`deepseek/deepseek-v4.1-flash`	37	`qwen/qwen-mt-lite`	1
OpenAI: gpt-oss-120b vs Qwen2.5-Coder-7B-Instruct	`openai/gpt-oss-120b`	37	`qwen/qwen2.5-coder-7b-instruct`	1
MoonshotAI: Kimi K2.6 vs Qwen3 30B A3B Instruct 2507	`moonshotai/kimi-k2.6`	27	`Qwen/Qwen3-30B-A3B-Instruct-2507`	2
Google: Gemma 4 31B vs Qwen3-30B-A3B-Thinking-2507	`google/gemma-4-31b-it`	26	`qwen/qwen3-30b-a3b-thinking-2507`	1
DeepSeek: DeepSeek V4 Pro 0423 vs Sao10K/L3.1-70B-Euryale-v2.2	`deepseek/deepseek-v4-pro`	26	`sao10k/l3.1-70b-euryale-v2.2`	2
MiniMax: MiniMax M3 vs euryale:33-70b	`minimax/minimax-m3`	25	`sao10k/l3.3-70b-euryale-v2.3`	1
Tencent: Hunyuan A13B Instruct vs Z.ai: GLM 5.3	`tencent/hunyuan-a13b-instruct`	2	`z-ai/glm-5.3`	50
TheDrummer: Cydonia 24B V4.1 vs Z.ai: GLM 5.2	`thedrummer/cydonia-24b-v4.1`	2	`z-ai/glm-5.2`	48
Z.ai: GLM 4.5 Air vs Z.ai: GLM 5.3 Flash	`z-ai/glm-4.5-air`	1	`z-ai/glm-5.3-flash`	48
MoonshotAI: Kimi K3 vs GLM 4.5 Air	`moonshotai/kimi-k3`	45	`zai-org/glm-4.5-air`	2
DeepSeek: DeepSeek V4 Flash 0731 vs GLM 4.6V	`deepseek/deepseek-v4-flash-0731`	35	`zai-org/glm-4.6v`	2
DeepSeek: DeepSeek V4.1 Flash vs @cf/ibm-granite/granite-4.0-h-micro	`deepseek/deepseek-v4.1-flash`	37	`ibm-granite/granite-4.0-h-micro`	1
@cf/aisingapore/gemma-sea-lion-v4-27b-it vs OpenAI: gpt-oss-120b	`aisingapore/gemma-sea-lion-v4-27b-it`	1	`openai/gpt-oss-120b`	37
ByteDance: UI-TARS 7B vs Qwen: Qwen3.8 27B	`bytedance/ui-tars-1.5-7b`	2	`qwen/qwen3.8-27b`	30
Cohere Embed v4.0 vs MoonshotAI: Kimi K2.6	`cohere/embed-v4.0`	2	`moonshotai/kimi-k2.6`	27
Google: Gemma 4 31B vs Inception: Mercury 2	`google/gemma-4-31b-it`	26	`inception/mercury-2`	1
DeepSeek: DeepSeek V4 Pro 0423 vs Llama 3.2 3B	`deepseek/deepseek-v4-pro`	26	`meta-llama/llama-3.2-3b`	2
Llama 3.3 70B vs MiniMax: MiniMax M3	`meta-llama/llama-3.3-70b`	2	`minimax/minimax-m3`	25