Beam 1.0: Benchmarks, Pricing & Context Window
Beam 1.0 is a language model from Reflection AI, released in October 2026, with a 1.0M-token context window, and pricing from $0.400/M input, $0.040/M cached input, $1.20/M output.
Beam 1.0 is Reflection AI's first open-weight model: a sparse Mixture-of-Experts reasoning model with 501.3B total parameters and 23.4B active per token (52 layers: 3 dense + 49 MoE; 256 experts; gated GQA with sliding-window attention;
Beam 1.0 benchmarks
Capability tiers
Standing within each category, adjusted for leaderboard depth.
Real tasks performance
High-confidence performance for Beam 1.0 across real-world prompt categories. Only 95% intervals at most 4 points wide are shown.
Performance by conversation depth
How Beam 1.0 holds up as conversations get longer.
Quality Tracker
Beam 1.0 Performance Across Datasets
Scores sourced from the model's scorecard, paper, or official blog posts
llm-stats.com - Wed Oct 21 2026Beam 1.0 pricing
Providers
Beam 1.0 starts at $0.400 per million input tokens and $1.20 per million output tokens via Together. Reused prompt prefixes cost $0.0400 per million cached input tokens.
| Provider | Input $/M | Cached input $/M | Output $/M | Context in / out | TTFT p95 s | Output p5 c/s | Modalities in / out |
|---|---|---|---|---|---|---|---|
 Together | $0.400 | $0.0400 | $1.20 | 1.0M/131.1K | — | — | / |
 Reflection AI | — | — | — | 1.0M/131.1K | — | — | / |
Cached input is the discounted price for prompt tokens served from a provider cache. TTFT is time to first token. Output is characters per second; p5 is the sustained floor exceeded by 95% of observed requests.
Beam 1.0 model size
Beam 1.0 has 501.3 billion parameters and was trained on 23.8 trillion tokens. See how it compares to other models in the same parameter range.
Beam 1.0 context window
Input and output token limits for Beam 1.0, plus how it ranks on long-context understanding.
Try now
huggleMake it with
Beam 1.0.
Beam 1.0 latency
Beam 1.0 time to first token, sustained output throughput, and failed-request rate from live model usage over the trailing 7 days.
Beam 1.0 examples
Recent arena outputs from Beam 1.0, picked from the highest-ranked matchups.
Beam 1.0 license
Beam 1.0 is released under the Apache 2.0 license, which permits commercial use, has 501.3B parameters, has a knowledge cutoff of June 2026.
- License
- Apache 2.0
- Commercial use allowed
- Parameters
- 501.3B
- Knowledge cutoff
- June 2026
Apache License 2.0 - allows commercial use
Beam 1.0 resources
Official sources for Beam 1.0: provider documentation, official playground, paper or system card, official launch post, model weights.
Beam 1.0 vs other models
The most-compared alternatives to Beam 1.0 are Claude Opus 4.6, GPT-5.2 Pro, Claude Opus 4.5. Open any pair side-by-side for benchmarks, pricing, context, and latency.
Models like Beam 1.0
Models ranked just above and below Beam 1.0 by LLM Stats score.
FAQ
Common questions about Beam 1.0.