Platform availability
as of 13 Oct 2026 · Token Factory public catalog
Model catalog
Newest first- DeepSeek-V4-Pro-0813
deepseek-ai
- GLM-5.3-Flash
zai-org
- DeepSeek-V4.1-Flash
deepseek-ai
- GLM-5.3
zai-org
- Qwen3.8-27B
Qwen
- Nemotron-3_5-Lightning
nvidia
- MiniMax-M3
MiniMaxAI
- DeepSeek-V4-Flash-0731
deepseek-ai
- Kimi-K3
moonshotai
- Kimi-K2.7-Code
moonshotai
- GLM-5.2
zai-org
- MiniCPM-V-4_5
openbmb
- Kimi-K2.6
moonshotai
- DeepSeek-V4-Pro
deepseek-ai
- GLM-5.1
zai-org
- Hermes-4-405B
NousResearch
- gpt-oss-120b
openai
- gemma-3-27b-it
google
Launch lag for flagship open-weight models
No overlap yet between catalog additions and tracked upstream releases.
Availability zones
Posted incidents, last 90 days
22 of 90 days16 Jul 2026darker = more severe · dashed = not covered13 Oct 2026
as of 13 Oct 2026 · Nebius status page
- 8 Oct 2026Metrics unavailable in Management Console for Monitoring in eu-west1major
- 8 Oct 2026Problems with power for several dozens of nodesmajor
- 7 Oct 2026Problems with power for several dozens of nodesmajor
- 5 Oct 2026Issues with preempted Managed Kubernetes nodes in all regionsminor
- 4 Oct 2026Issues with preempted Managed Kubernetes nodes in all regionsminor
- 3 Oct 2026InfiniBand connectivity issues for GPU clusters in eu-west2minor
- 2 Oct 2026Compute VM creation and Object storage are partial unavailablemajor
Distribution through OpenRouter
as of 13 Oct 2026 · OpenRouter public models API
| OpenRouter model | Nebius | Nebius $/M out | price rank | cheapest $/M out | quant. | uptime 1d | src |
|---|---|---|---|---|---|---|---|
| google/gemma-3-27b-it | yes | $0.300 | 4 / 5 | $0.160 | fp8 | 94.8% | Open source |
| moonshotai/kimi-k2.7-code | yes | $4.00 | 7 / 12 | $3.00 | fp4 | 98.9% | Open source |
| nousresearch/hermes-4-405b | yes | $3.00 | 1 / 1 | $3.00 | fp8 | 99.9% | Open source |
| openai/gpt-oss-120b | yes | $0.600 | 13 / 24 | $0.150 | fp4 | 99.5% | Open source |
| qwen/qwen3-235b-a22b-2507 | yes | $0.600 | 3 / 7 | $0.350 | fp8 | 77.2% | Open source |
| qwen/qwen3-30b-a3b-instruct-2507 | yes | $0.300 | 1 / 3 | $0.300 | fp8 | 89.1% | Open source |
| z-ai/glm-5.1 | yes | $4.40 | 7 / 13 | $3.04 | fp8 | 98.0% | Open source |
| z-ai/glm-5.2 | yes | $4.40 | 14 / 34 | $1.68 | fp4 | 97.8% | Open source |
| deepseek/deepseek-v4-flash-0731 | no | n/a | – / 27 | $0.180 | n/a | n/a | Open source |
| deepseek/deepseek-v4-pro | no | n/a | – / 15 | $1.91 | n/a | n/a | Open source |
| deepseek/deepseek-v4-pro-0813 | no | n/a | – / 22 | $1.28 | n/a | n/a | Open source |
| deepseek/deepseek-v4.1-flash | no | n/a | – / 32 | $0.180 | n/a | n/a | Open source |
| minimax/minimax-m3 | no | n/a | – / 13 | $0.960 | n/a | n/a | Open source |
| moonshotai/kimi-k2.6 | no | n/a | – / 17 | $1.95 | n/a | n/a | Open source |
| moonshotai/kimi-k3 | no | n/a | – / 25 | $9.50 | n/a | n/a | Open source |
| qwen/qwen3.5-397b-a17b | no | n/a | – / 10 | $2.34 | n/a | n/a | Open source |
| qwen/qwen3.8-27b | no | n/a | – / 20 | $0.500 | n/a | n/a | Open source |
| z-ai/glm-5.3 | no | n/a | – / 42 | $2.20 | n/a | n/a | Open source |
| z-ai/glm-5.3-flash | no | n/a | – / 32 | $0.050 | n/a | n/a | Open source |
Price rank 1 is the cheapest output price among the model's OpenRouter providers. Uptime is OpenRouter's own measurement of its traffic to the endpoint. How to read price rank and uptime.