model comparison

vs

Beam vs Claude Haiku 5.5

Reflection against Anthropic, compared on listed price, real workload cost, context and capabilities.

the short answer

  • Claude Haiku 5.5 lists the lower input price, while Beam lists the lower output price.
  • Claude Haiku 5.5 supports vision; Beam does not.
  • Claude Haiku 5.5 reads cached input for less ($0.01/M against Beam's $0.015/M), which decides the bill for agents that resend long context.

side-by-side specs

specBeamClaude Haiku 5.5
providerReflectionAnthropic
input price$0.125/M$0.1/M
output price$0.4/M$0.5/M
cached input$0.015/M$0.01/M
blended price (3:1)$0.194/M$0.2/M
context window1M tokens1M tokens
max output128k tokens128k tokens
inputtexttext, images
structured outputyesyes
weightsopen (Apache 2.0)closed
releasedOctober 2026October 7, 2026

same work, both models

workloadBeamClaude Haiku 5.5
one chat reply3,000 tokens in, 500 outwithin 5%$0.00058$0.00055
summarise a 100-page report50,000 tokens in, 1,500 out, one requestBeam costs 1.2× as much as Claude Haiku 5.5$0.0069$0.0058
1,000 classification calls500 tokens in, 20 out, per call$0.07$0.06
coding-agent session2M tokens in across ~60 turns, 60,000 out, no cache$0.27$0.23
same session, cached85% of input read from the prompt cache, the rest written to itwithin 5%$0.09$0.08

at listed rates. token counts are typical sizes, not measurements.

Claude Haiku 5.5 pricing notes

  • every rate rises 5x once a prompt passes 100K tokens, counting cache reads and writes; the listed price is the sub-100K rate.
  • cache writes bill at $0.125 and cache hits at $0.01 per 1M tokens below 100K prompt tokens.
  • its tokenizer counts the same text as about 30% more tokens than Claude Haiku 4.5.

when to pick which

Beam

the first open-weights model from Reflection, a 501B-total, 23B-active mixture-of-experts built for agentic tool use, coding and general reasoning. it reads text only and its weights ship under Apache 2.0.

pick it if

  • you want a long-window text model priced with the flash tier
  • you want Apache 2.0 weights, with BF16, FP8 and NVFP4 builds to self-host
  • you run tool-calling agents on text and repeat long prompts, so cached input matters

skip it if

  • you need image, audio or video input; Beam reads text only
  • long autonomous coding is the job; its reported DeepSWE v1.1 score sits below GLM-5.3 Flash and DeepSeek V4.1 Flash

Claude Haiku 5.5

Anthropic's small Claude model, with adaptive thinking, image input and a 1M-token context. it is the low-cost tier below Sonnet 5.5, priced at a tenth of Haiku 4.5's rates for prompts up to 100K tokens.

pick it if

  • you run high-volume classification, extraction, routing or subagent tasks
  • your prompts stay under 100K tokens, where every rate is lowest
  • you resend long prompts, so cache hits bill at a tenth of input

skip it if

  • you need complex agentic coding; Anthropic points to Sonnet 5.5 or Opus 5.5
  • your prompts routinely pass 100K tokens, where every rate is 5x

faq

which is cheaper, Beam or Claude Haiku 5.5?

Claude Haiku 5.5 lists the lower input price, while Beam lists the lower output price. on a coding-agent session (2M tokens in, 60,000 out) Beam costs $0.27 and Claude Haiku 5.5 $0.23.

which one accepts images?

Claude Haiku 5.5 supports vision; Beam does not.

can i switch between Beam and Claude Haiku 5.5 mid-task?

yes. in an octomind session, /model octohub:beam and /model octohub:claude-haiku-5-5 switch models without losing the conversation, so you can draft on one and review on the other.

try either model

switch between models inside the same octomind session.

open the panel all models

prices from the catalog of 2026-11-09