Roboflow

Claude Haiku 5.5 vs Qwen3.8 27B

Claude Haiku 5.5 averages 69.6% across five Vision Evals tasks and Qwen3.8 27B 69.5% across four (#31 and #32 of 73 models), so they are effectively tied.

  • Claude Haiku 5.5 wins 3 of the 4 tasks both ran, with its widest lead on reasoning (68.9% vs 62.0%).
  • Claude Haiku 5.5 is about half the cost: $0.0006 vs $0.0009 overall average cost per sample.
  • Claude Haiku 5.5 is faster: 11.0s vs 18.0s per sample.

Compare Claude Haiku 5.5 vs Qwen3.8 27B live

Run the same image across every model that supports a task and compare their outputs side-by-side.

Detect and compare bounding boxes across models on the same image.

Open Object Detection in the full playground
AnthropicClaude Haiku 5.5
Run to compare this model.
QwenQwen3.8 27B
Run to compare this model.

Claude Haiku 5.5 vs Qwen3.8 27B on Vision Evals

Score by taskClaude Haiku 5.5Qwen3.8 27B
Reasoning
68.9
62.0
Counting
68.9
64.9
Identification
83.3
85.4
Object Detection
65.8
65.7

Roboflow Vision Evals. Full breakdown below.

Claude Haiku 5.5 vs Qwen3.8 27B Comparison Table

Evals updated November 19, 2026Pricing updated November 22, 2026

PropertyClaude Haiku 5.5Qwen3.8 27B
OrganizationAnthropicQwen
Categoryclosedopen
Modalitymultimodalmultimodal
Release DateOct 2026Aug 2026
Context Window1.0M262K
Parametersundisclosed27.78B
LicenseProprietaryApache 2.0
Commercial usecommercial use per vendor termsfree for commercial use
Pricing per 1M tokens
Input $/1M$0.100$0.035
Output $/1M$0.500$1.50
Vision Tasks
Both support 10 more tasks
Model Features
Both support 3 more features
Vision Evalsground-truth scores across 5 vision tasks, pooled at low effort
Overall
69.6%
69.5%*
Quantizationsself-hosted
BF1669.5%FP868.7%AWQ-INT469.5%hardware →
Avg cost / sample$0.0006$0.0009
Avg speed / sample11.01s17.99s
By task
Object Detection (low)
65.8%
±0.6, Mean of 3 runs, range 65.1 to 66.2
$0.0006
65.7%
±1.0, Mean of 3 runs, range 64.6 to 66.5
$0
Object Detection (high)
68.2%
±1.0, Mean of 3 runs, range 67.2 to 69.2
$0.0010
66.1%
±1.4, Mean of 3 runs, range 64.9 to 67.8
$0
Counting (low)
68.9%
±4.7, Mean of 3 runs, range 64.9 to 74.3
$0.0003
64.9%
±4.1, Mean of 3 runs, range 60.8 to 68.9
$0
Counting (high)
73.0%
±1.3, Mean of 3 runs, range 71.6 to 74.3
$0.0004
68.0%
±2.0, Mean of 3 runs, range 66.2 to 70.3
$0
Identification (low)
83.3%
±3.1, Mean of 3 runs, range 81.3 to 87.5
$0.0002
85.4%
±4.7, Mean of 3 runs, range 81.3 to 90.6
$0
Identification (high)
86.5%
±1.6, Mean of 3 runs, range 84.4 to 87.5
$0.0003
87.5%
±3.1, Mean of 3 runs, range 84.4 to 90.6
$0
OCR (low)
61.2%
$0.0006
–
by category
Single value
44.8%
Transcription
93.0%
Structured JSON
82.2%
Text localization
32.0%
OCR (high)
65.3%
$0.0010
–
by category
Single value
46.5%
Transcription
94.7%
Structured JSON
85.3%
Text localization
45.8%
Reasoning (low)
68.9%
±2.0, Mean of 3 runs, range 66.9 to 70.9
$0.0004
62.0%
±2.0, Mean of 3 runs, range 60.3 to 64.2
$0
Reasoning (high)
74.8%
±2.6, Mean of 3 runs, range 72.2 to 77.5
$0.0006
66.0%
±0.7, Mean of 3 runs, range 65.6 to 66.9
$0

Claude Haiku 5.5 vs Qwen3.8 27B: Overview

Claude Haiku 5.5

Claude Haiku 5.5 is a proprietary multimodal language model from Anthropic and the smallest member of the Claude 5.5 family, released on October 7, 2026 after Claude Opus 5.5 and Claude Sonnet 5.5. It accepts text and image input and returns text, with a 1M token context window and up to 128K output tokens per request. It is the first Haiku-class model with an adjustable effort parameter: adaptive thinking is on by default and the model decides how much to reason, steered by effort levels from low to max with medium as the default. Its training data cutoff is June 2026. Pricing starts at $0.10 per million input tokens and $0.50 per million output tokens for prompts up to 100K tokens, which Anthropic reports is about 90% lower than Claude Haiku 4.5 for requests in that range.

Anthropic positions Haiku 5.5 for high-volume, latency-sensitive work such as classification, extraction, routing, summarization, and subagent tasks, and describes it as its fastest model to date at standard speed. It uses the same tokenizer as Claude Opus 4.7 and later models, so the same text counts as roughly 30% more tokens than on Haiku 4.5.

Qwen3.8 27B

Qwen3.8-27B is a dense vision-language model of roughly 27.78 billion parameters from Alibaba's Qwen team, built on the architectural foundation established by the Qwen3.5 generation. It takes interleaved text, image, and video input through an integrated vision encoder and produces text, covering document and chart understanding, visual question answering, front-end code generation from visual references, and agentic computer-use workflows. The decoder is a hybrid stack of 64 layers that repeats a pattern of three Gated DeltaNet linear-attention blocks, each followed by a feed-forward block, then one grouped-query full-attention block, so only a quarter of the layers maintain a conventional key-value cache. Native context length is 262,144 tokens.

Post-training targets long-horizon agentic work, and the model exposes per-request thinking control that lets callers switch reasoning traces on or off and choose a reasoning effort level. Multi-token prediction weights are trained alongside the main network for speculative decoding.

Frequently Asked Questions

On Roboflow's Vision Evals, Claude Haiku 5.5 performed better. It scores higher on 3 of the 4 vision tasks and averages 69.6% (#31 of 73) against 69.5% (#32 of 73) for Qwen3.8 27B. These scores measure vision capabilities only; either model may perform differently on work outside this benchmark's tasks.

Yes. On the Vision Evals Reasoning benchmark at low effort, Claude Haiku 5.5 leads with 68.9% against 62.0%. This is the widest gap between the two models across the benchmark's tasks.

Claude Haiku 5.5 is cheaper to run. Across Roboflow's Vision Evals task mix it averaged $0.0006 per sample against $0.0009. Actual costs depend on your image sizes, prompts, and output length.

Claude Haiku 5.5 is faster. Across Roboflow's Vision Evals it averaged 11.0s per inference against 18.0s. Latency varies with image size, prompt length, and provider load, so treat these as relative rather than guaranteed figures.

Yes. The comparison demo on this page runs both models on the same image side by side for image captioning and OCR in the free Roboflow Playground. You can try it instantly, and a free account gives you more runs with free Roboflow credits.