Models

Runtime registry with capability flags, context limits, pricing, and route state.

3 models indexed

Gemma 3 27B

google-gemma-3-27bgemmaGoogle

warm
toolsvisionjson_modestreamingcompletion
Context: 200,000
Max output: 25,000
Pricing: 0.7/6.7 per 1M
Class: closed
Model detail + compatibility

Gemma 3 27B listed under llm foundation models for AI Bazaar discovery and comparison workflows.

temperature: supported
top_p: supported
top_k: supported
min_p: supported
max_tokens: supported
frequency_penalty: supported
presence_penalty: supported
stop: supported
seed: supported
tools: supported
vision: supported
stream: supported
response_format_json: supported
curl -X POST /api/v1/chat/completions -d '{"model":"google-gemma-3-27b"}'

Gemma 3 27B Instruct

google-gemma-3-27b-itgemmaGoogle

warm
toolsjson_modestreamingcompletion
Context: 131,072
Max output: 16,384
Pricing: 0.16/0.5 per 1M
Class: open
Model detail + compatibility

Open-weight instruction model optimized for cost-efficient coding and multilingual tasks.

temperature: supported
top_p: supported
top_k: supported
min_p: supported
max_tokens: supported
frequency_penalty: supported
presence_penalty: supported
stop: supported
seed: supported
tools: supported
vision: not supported
stream: supported
response_format_json: supported
curl -X POST /api/v1/chat/completions -d '{"model":"google-gemma-3-27b-it"}'

Gemma Code 2

google-gemma-code-2gemmaGoogle

warm
toolsstreamingcompletion
Context: 200,000
Max output: 8,192
Pricing: 0.45/1 per 1M
Class: closed
Model detail + compatibility

Gemma Code 2 listed under coding models for AI Bazaar discovery and comparison workflows.

temperature: supported
top_p: supported
top_k: supported
min_p: supported
max_tokens: supported
frequency_penalty: supported
presence_penalty: supported
stop: supported
seed: supported
tools: supported
vision: not supported
stream: supported
response_format_json: not supported
curl -X POST /api/v1/chat/completions -d '{"model":"google-gemma-code-2"}'
⌘K
AI Bazaar