Gemini Flashgemini-flash
Multimodal speed with a large context
- Context
- 1,000,000 tokens
- Provider list price
- $0.1 in / $0.4 out per 1M
- Upstream
- google/gemini-2.5-flash
StreamingVisionFilesToolsWeb search
Model catalog
Models
The models Umbra can route to, with provider list prices and the capabilities declared in the local catalog.
Qwen Coderqwen-coder
Focused help for code and debugging
- Context
- 262,144 tokens
- Provider list price
- $0.3 in / $1 out per 1M
- Upstream
- qwen/qwen3-coder
StreamingFilesTools
Sage Sonnetsage-sonnet
Careful writing and analysis
- Context
- 200,000 tokens
- Provider list price
- $3 in / $15 out per 1M
- Upstream
- anthropic/claude-sonnet-4.5
StreamingVisionFilesTools
Llama Openllama-open
Open-weight conversation
- Context
- 131,072 tokens
- Provider list price
- $0.4 in / $0.4 out per 1M
- Upstream
- meta-llama/llama-3.3-70b-instruct
StreamingTools
Umbra Autoumbra-auto
A balanced route selected for your prompt
- Context
- 128,000 tokens
- Provider list price
- $0.15 in / $0.6 out per 1M
- Upstream
- openai/gpt-4o-mini
StreamingVisionFilesToolsWeb search
Nova 4nova-4
Fast, capable everyday reasoning
- Context
- 128,000 tokens
- Provider list price
- $2.5 in / $10 out per 1M
- Upstream
- openai/gpt-4o
StreamingVisionFilesTools
Reasoning R1reasoning-r1
Deep, deliberate problem solving
- Context
- 65,536 tokens
- Provider list price
- $0.55 in / $2.19 out per 1M
- Upstream
- deepseek/deepseek-r1
StreamingReasoning
Mistral Smallmistral-small
Efficient and precise
- Context
- 32,000 tokens
- Provider list price
- $0.1 in / $0.3 out per 1M
- Upstream
- mistralai/mistral-small-3.1-24b-instruct
StreamingVisionTools