A reactive notebook for Python — run reproducible experiments, query with SQL, execute as a script, deploy as an app, and version with git. Stored as pure Python. All in a modern, AI-native editor.
Python artificial-intelligencedagdata-sciencedata-visualizationdataflowdeveloper-toolsmachine-learningnotebookspipelinepythonreactivesqlweb-app
Update LLM catalog with Beam and DeepSeek Flash 1017 (#11315)
github-actions[bot]8ee7ff1a6256
diff --git a/packages/llm-info/data/models.yml b/packages/llm-info/data/models.ymlindex f3390211a4..a1a94391b5 100644--- a/packages/llm-info/data/models.yml+++ b/packages/llm-info/data/models.yml@@ -1141,6 +1141,21 @@ azure:limits: {context: 1000000, output: 384000}ollama:+ - name: DeepSeek V4.1 Flash 1017+ model: deepseek-v4.1-flash:1017+ description: "October 17 update of DeepSeek V4.1 Flash for coding and end-to-end project development, with faster task completion and shorter outputs"+ roles: [chat, edit]+ capabilities: [thinking, tool_calling]+ input_types: [text, image]+ output_types: [text]+ release_date: 2026-10-17+ cost: {input: 0.15, output: 0.6}+ reasoning_options:+ - type: toggle+ - type: effort+ values: [low, high, max]+ limits: {context: 1048576, output: 384000}+- name: Muse Spark Openmodel: muse-spark-opendescription: "Open-weight model optimized for agentic coding, reliable tool calling, and codebase understanding"@@ -1382,6 +1397,20 @@ ollama:cost: {input: 0.6, output: 3.6}wandb:+ - name: Beam 1.0+ model: reflection/Beam-1.0-501B-A23B+ description: "Open-weight reasoning model from Reflection for coding and long-running agent workflows, with 501B total and 23B active parameters"+ roles: [chat, edit]+ capabilities: [thinking, tool_calling]+ input_types: [text]+ output_types: [text]+ release_date: 2026-10-16+ cost: {input: 1, output: 3}+ reasoning_options:+ - type: effort+ values: [minimal, low, medium, high, xhigh, max]+ limits: {context: 1048576, output: 1048576}+- name: Nemotron 3.5 Super VLmodel: nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12Bdescription: "Open multimodal Mixture-of-Experts model with 12B active parameters out of 120B for reasoning, coding, tool use, and document understanding"@@ -1390,7 +1419,7 @@ wandb:input_types: [text, image]output_types: [text]release_date: 2026-10-15- cost: {input: 0.15, output: 0.75}+ cost: {input: 0.15, output: 0.5}reasoning_options:- type: togglelimits: {context: 1048576, output: 1048576}@@ -1595,6 +1624,20 @@ wandb:limits: {context: 131072, output: 131072}opencode-go:+ - name: Beam Free+ model: beam-free+ description: "Free access to Reflection's Beam model for coding, reasoning, and long-running agent workflows, available for a limited time in supported regions"+ roles: [chat, edit]+ capabilities: [thinking, tool_calling]+ input_types: [text]+ output_types: [text]+ release_date: 2026-10-05+ cost: {input: 0, output: 0}+ reasoning_options:+ - type: effort+ values: [low, medium, high, xhigh, max]+ limits: {context: 1048576, output: 131072}+- name: Gemini 4 Argon Contributormodel: gemini-4-argon-contributordescription: "Contributor tier of Gemini 4 Argon for reasoning, coding, and agents, with prompts and completions eligible for model training"