Beam 1.0
Freevia Poolreflection/beam-1.0262K context · Oct 2026 · 1 upstreamsupstream上游
Beam 1.0 is Reflection's open-weights mixture-of-experts model — 501B total parameters, 23B active across 256 experts — built for token-efficient reasoning, agentic tool use, and software-engineering workloads. Reasoning is always on with a tunable effort level; the Apache-2.0 weights are on Hugging Face.
- contextngữ cảnh上下文
- 262K
- inputđầu vào输入 / 1M
- $0
- outputđầu ra输出 / 1M
- $0
- routestuyến路由
- 1
- speed, 24htốc độ, 24h速度,24h
- —
Upstreams
1- 1Nous ResearchPlatform · freeFreeTPS —TTFT ——
- Up
- No traffic
- Down
Try itDùng thử试用
Send a promptGửi một prompt发送提示词
Details
- Owner
- Reflection
- Kind
- Chat
- Released
- Oct 2026
- Context
- 262K
- Data policy
- Provider retention
- Endpoint
/api/v1/chat/completions
Quick start
curl https://anyrouter.dev/api/v1/chat/completions \ -H "Authorization: Bearer $ANYROUTER_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "reflection/beam-1.0", "messages": [ { "role": "user", "content": "Say hi in 3 words." } ] }'
- Agentic coding: real GitHub issues, CLI and DevOps work
- Long-horizon tool workflows over MCP
- Math, science, and applied reasoning
Architecture and licenseKiến trúc và giấy phép架构与许可
- TypeLoại类型
- Mixture of experts[1]
- Total parametersTổng tham số总参数
- 501.3B[1]
- Active parametersTham số kích hoạt激活参数
- 23.4B[1]
- LicenseGiấy phép许可
- Apache-2.0 · Open weightsTrọng số mở开放权重[1]
StrengthsĐiểm mạnh优势[1]
- Token-efficient reasoning — comparable quality with ~3x fewer reasoning tokens
- Agentic tool use: multi-step workflows, MCP tool integration, browsing
- Strong SWE scores: 81.6% SWE-bench Verified, 80.1% Terminal Bench v2.1
Known limitsGiới hạn đã biết已知局限[1],[2]
- Developed predominantly for English
- Reasoning is mandatory — there is no non-reasoning mode
- Hosts serve a 262144-token window, below the model's 1M training context
API notesGhi chú APIAPI 说明[1],[2]
- reasoning_effort accepts low/medium/high/xhigh/max; default max, cannot be turned off
- Chat template is a modified Harmony format
- Tool calling and structured outputs (JSON schema) supported
Links and versionsLiên kết và phiên bản链接与版本
- AnnouncementBài công bố发布公告
- reflection.ai
- Model cardThẻ mô hình模型卡
- huggingface.co
- DocsTài liệu文档
- developers.reflection.ai
SourcesNguồn来源
- [1]reflection/Beam-1.0-501B-A23B model cardaccessedtruy cập访问于 2026-10-17
- [2]Reflection API — Modelsaccessedtruy cập访问于 2026-10-17
- [3]Cloudflare Workers AI — Beam 1.0 501B A23Baccessedtruy cập访问于 2026-10-17
Upstreams and pricingUpstream và giá上游与价格
Share cards3 images



AliasesBí danh别名 (2)
reflection-ai/beam-1.0-501b-a23breflection/beam-1.0-501b-a23b
params: max_tokens · temperature · top_p · top_k · tools · tool_choice · response_format · reasoning · include_reasoning · reasoning_effort
Uptime + latency
—recent checks · all routes
No health checks recorded for this model yet.
24h
No traffic in the last 24h.