Kontinent
Sign in

kontinent / status

Measured model availability

We call every served chat model every 30 minutes with the same prompt, through our own gateway. What you see here is the result of those calls — not a third-party report and not an estimate.

ProviderAvailableResponse
AKI.IO100.0 %884 ms
Apertus 70B100.0 %1.4 s
DeepSeek V4.1 Flash (AKI.IO)100.0 %896 ms

1 of 5 calls served through the fallback chain: tensorx/deepseek-v4.1-flash

Gemma 4 26B (AKI.IO)100.0 %977 ms
GLM 5.3 (AKI.IO)100.0 %770 ms
GPT-OSS 120B (AKI.IO)100.0 %366 ms

1 of 36 calls served through the fallback chain: ovh/gpt-oss-120b

Mistral 4 119B100.0 %436 ms
Qwen3.6 35B (AKI.IO)100.0 %1.1 s

1 of 36 calls served through the fallback chain: scaleway/qwen3.6-35b

Qwen3.8 27B (AKI.IO)100.0 %602 ms
AWS Bedrock (EU Frankfurt)99.9 %760 ms
Claude Opus 4.8 (Bedrock)97.2 %847 ms
Claude Haiku 4.5 (Bedrock)100.0 %671 ms
Claude Haiku 5.5 (Bedrock)100.0 %779 ms
Claude Opus 4.5 (Bedrock)100.0 %1.4 s
Claude Opus 4.6 (Bedrock)100.0 %1.4 s
Claude Opus 4.7 (Bedrock)100.0 %837 ms
Claude Opus 5 (Bedrock)100.0 %1.0 s
Claude Opus 5.5 (Bedrock)100.0 %1.3 s
Claude Sonnet 4.5 (Bedrock)100.0 %1.5 s
Claude Sonnet 4.6 (Bedrock)100.0 %828 ms
Claude Sonnet 5 (Bedrock)100.0 %1.1 s
Claude Sonnet 5.5 (Bedrock)100.0 %905 ms
Devstral 2 123B (Bedrock)100.0 %322 ms
GLM 4.7 Flash (Bedrock)100.0 %283 ms
GPT-OSS 120B (Bedrock)100.0 %491 ms
GPT-OSS 20B (Bedrock)100.0 %402 ms
MiniMax M2.1 (Bedrock)100.0 %604 ms
MiniMax M2.5 (Bedrock)100.0 %570 ms
Nova 2 Lite (Bedrock)100.0 %462 ms
Nova Lite (Bedrock)100.0 %531 ms
Nova Micro (Bedrock)100.0 %539 ms
Nova Pro (Bedrock)100.0 %691 ms
Pixtral Large (25.02) (Bedrock)100.0 %701 ms
Qwen3 235B A22B 2507 (Bedrock)100.0 %294 ms
Qwen3 32B (dense) (Bedrock)100.0 %303 ms
Qwen3-Coder-30B-A3B-Instruct (Bedrock)100.0 %294 ms
AWS Bedrock (EU Irland)100.0 %462 ms
Gemma 3 12B (Bedrock)100.0 %430 ms
Gemma 3 27B (Bedrock)100.0 %417 ms
Gemma 3 4B (Bedrock)100.0 %390 ms
Magistral Small (Bedrock)100.0 %534 ms
MiniMax M2 (Bedrock)100.0 %1.2 s
Ministral 3 14B (Bedrock)100.0 %406 ms
Ministral 3 3B (Bedrock)100.0 %405 ms
Ministral 3 8B (Bedrock)100.0 %439 ms
NVIDIA Nemotron Nano 12B v2 (Bedrock)100.0 %352 ms
NVIDIA Nemotron Nano 3 30B (Bedrock)100.0 %476 ms
NVIDIA Nemotron Nano 9B v2 (Bedrock)100.0 %1.3 s
Qwen3 Next 80B A3B (Bedrock)100.0 %357 ms
Qwen3-VL 235B A22B (Bedrock)100.0 %472 ms
Azure OpenAI (EU Data Zone)99.8 %775 ms
GPT-4.1 (Azure EU)97.2 %751 ms
GPT-4.1 mini (Azure EU)100.0 %470 ms
GPT-4.1 nano (Azure EU)100.0 %441 ms
GPT-4o mini (Azure EU)100.0 %380 ms
GPT-5 (Azure EU)100.0 %2.4 s
GPT-5 mini (Azure EU)100.0 %956 ms
GPT-5 nano (Azure EU)100.0 %992 ms
GPT-5.1 (Azure EU)100.0 %666 ms
GPT-5.4 (Azure EU)100.0 %642 ms
GPT-5.4 mini (Azure EU)100.0 %557 ms
GPT-5.5 (Azure EU)100.0 %773 ms
GPT-5.6 Luna (Azure EU)100.0 %578 ms
GPT-5.6 Sol (Azure EU)100.0 %735 ms
GPT-5.6 Terra (Azure EU)100.0 %598 ms
GPT-6 Luna (Azure EU)100.0 %653 ms
GPT-6 Sol (Azure EU)100.0 %1.2 s
GPT-6.1 Sol (Azure EU)100.0 %1.8 s
o3 (Azure EU)100.0 %1.1 s
o4-mini (Azure EU)100.0 %934 ms
Google Vertex AI (EU)100.0 %1.2 s
Gemini 2.5 Flash (Vertex)100.0 %992 ms
Gemini 2.5 Flash-Lite (Vertex)100.0 %516 ms
Gemini 2.5 Pro (Vertex)100.0 %4.4 s
Gemini 3.1 Flash-Lite (Vertex)100.0 %678 ms
Gemini 3.5 Flash (Vertex)100.0 %1.2 s
Gemini 3.5 Flash-Lite (Vertex)100.0 %634 ms
Gemini 3.6 Flash (Vertex)100.0 %1.2 s
Gemini 3.7 Flash (Vertex)100.0 %1.9 s
Gemini 3.8 Flash (Vertex)100.0 %1.8 s
Inceptron99.6 %1.2 s
Kimi K2.6 (Inceptron)97.2 %1.3 s

1 of 36 calls served through the fallback chain: tensorx/kimi-k2.6

DeepSeek V4 Flash (Inceptron)100.0 %925 ms
GLM 5.2 (Inceptron)100.0 %744 ms
GLM 5.3 (Inceptron)100.0 %1.4 s
GLM 5.3 Flash (Inceptron)100.0 %2.2 s

2 of 36 calls served through the fallback chain: tensorx/glm-5.3-flash

Kimi K2.7-Code 1100B (Inceptron)100.0 %652 ms
Infercom100.0 %808 ms
Gemma 4 31B (Infercom)100.0 %939 ms
GPT-OSS 120B (Infercom)100.0 %370 ms
IONOS AI Model Hub99.7 %416 ms
Llama 3.1 8B (IONOS)97.2 %109 ms
GPT-OSS 120B (IONOS)100.0 %327 ms
Llama 3.3 70B (IONOS)100.0 %181 ms
Mistral NeMo (IONOS)100.0 %355 ms
Mistral Small 24B (IONOS)100.0 %312 ms
Qwen3 Coder Next (IONOS)100.0 %137 ms
Qwen3.5 397B (IONOS)100.0 %2.7 s
Qwen3.5 9B (IONOS)100.0 %2.9 s
Qwen3.8 27B (IONOS)100.0 %742 ms
Mistral La Plateforme100.0 %323 ms
Codestral100.0 %249 ms
Ministral 3 14B100.0 %564 ms

1 of 36 calls served through the fallback chain: bedrock/ministral-3-14b

Ministral 3 3B100.0 %287 ms
Ministral 3 8B100.0 %346 ms
Mistral Medium 3.5100.0 %330 ms

36 of 36 calls served through the fallback chain: scaleway/mistral-medium-3.5

Mistral Small 4100.0 %245 ms

36 of 36 calls served through the fallback chain: regolo/mistral-small-4

Nebius Token Factory100.0 %438 ms
Gemma 3 27B (Nebius)100.0 %259 ms
GLM-5.1 (Nebius)100.0 %2.2 s
GPT-OSS 120B (Nebius)100.0 %272 ms
Hermes 4 405B (Nebius)100.0 %275 ms
Kimi K3 (Nebius)100.0 %1.2 s
MiniCPM-V 4.5 (Nebius)100.0 %117 ms
Nemotron 3 Nano 30B (Nebius)100.0 %493 ms
Nemotron 3.5 Lightning (Nebius)100.0 %472 ms
Qwen3 235B A22B (Nebius)100.0 %342 ms
Qwen3 30B A3B (Nebius)100.0 %223 ms
Nebul100.0 %434 ms
GLM-5.3 (Nebul)100.0 %695 ms
GPT-OSS 120B (Nebul)100.0 %353 ms
Ministral 3 14B (Nebul)100.0 %161 ms
Mistral Large 3 (Nebul)100.0 %172 ms
Nemotron 3 Super 120B (Nebul)100.0 %399 ms
Qwen3 30B A3B (Nebul)100.0 %474 ms
Qwen3.5 397B (Nebul)100.0 %1.1 s
OVHcloud AI Endpoints100.0 %1.0 s
GPT-OSS 120B (OVHcloud)100.0 %989 ms
GPT-OSS 20B100.0 %734 ms
Llama 3.3 70B (OVHcloud)100.0 %188 ms
Mistral NeMo (OVHcloud)100.0 %451 ms
Mistral Small 3.2 (OVHcloud)100.0 %364 ms
Qwen2.5 VL 72B100.0 %161 ms
Qwen3.5 397B (OVHcloud)100.0 %1.8 s
Qwen3.5 9B100.0 %4.6 s
Qwen3.6 27B100.0 %1.8 s
Qwen3.8 27B (OVHcloud)100.0 %2.8 s
Regolo AI100.0 %773 ms
Apertus 70B (Regolo)100.0 %932 ms
Gemma 4 31B (Regolo)100.0 %126 ms
GLM 5.2 (Regolo)100.0 %2.6 s
GPT-OSS 120B (Regolo)100.0 %201 ms
GPT-OSS 20B (Regolo)100.0 %248 ms
Mistral Small 4 (Regolo)100.0 %74 ms
Qwen3.5 122B (Regolo)100.0 %977 ms
Qwen3.5 9B (Regolo)100.0 %1.0 s
Qwen3.8 27B (Regolo)100.0 %680 ms
Scaleway Generative APIs100.0 %433 ms
DeepSeek V4 Flash (Scaleway)100.0 %455 ms
Gemma 4 26B100.0 %373 ms
GLM 5.2100.0 %1.2 s
GPT-OSS 120B100.0 %394 ms
Llama 3.3 70B (Scaleway)100.0 %120 ms
Mistral Medium 3.5 (Scaleway)100.0 %123 ms
Mistral Small 3.2 (Scaleway)100.0 %164 ms
Qwen3 235B100.0 %102 ms
Qwen3.5 397B100.0 %1.9 s
Qwen3.6 35B100.0 %1.3 s
Qwen3.8 27B (Scaleway)100.0 %304 ms
TensorX100.0 %858 ms
DeepSeek V4 Flash (0731) (TensorX)100.0 %418 ms
DeepSeek V4 Pro 08.13100.0 %322 ms
DeepSeek V4.1 Flash100.0 %488 ms
GLM 5.2 (TensorX)100.0 %856 ms
GLM 5.3 (TensorX)100.0 %1.3 s
GLM 5.3 Flash (TensorX)100.0 %2.1 s

1 of 40 calls served through the fallback chain: inceptron/glm-5.3-flash

Kimi K2.7-Code 1100B (TensorX)100.0 %416 ms
Kimi K3100.0 %2.0 s
MiniMax M3100.0 %751 ms
Qwen3.8 2.4T100.0 %398 ms
Qwen3.8 27B100.0 %1.5 s
Qwen3.8 Flash Next100.0 %508 ms

How we measure

The source is our own probe traffic only, never customer traffic. Every 30 minutes the same short prompt goes to every served chat model, plus one realistically long prompt once a day. Each bar is one day. “Available” is the share of successful calls over those same seven days; “Response” is the median of successful calls. A grey bar means we did not measure that day — not that the model was down. A provider takes the colour of its worst model.

When a model fails, the gateway tries the fallback chain behind it. The call still counts as available, because the customer did get an answer. The response time then includes the failed attempt. Where this happened it is noted under the model row, naming the model that actually answered. A green row carrying that note means: usable, but not answered by this provider.

This page describes the availability of individual models and providers. It is not a statement about the platform availability committed in § 7 of the Terms of Service — we expressly do not warrant any particular availability of individual models.