Anthropic: Claude 3 Haiku comes out ahead on the data we track: 3 advantages and none for unsloth/Qwen3.8-27B-NVFP4.
| Spec | Anthropic: Claude 3 Haiku | unsloth/Qwen3.8-27B-NVFP4 |
|---|---|---|
| Provider | Anthropic | Hugging Face |
| Input price (per 1M tokens) | $0.25 | n/a |
| Output price (per 1M tokens) | $1.25 | n/a |
| Blended price (3:1 input/output) | $0.50 | n/a |
| Cheaper than … of all models we track | 53% | n/a |
| Context window | 200K tokens (~300 pages) | n/a |
| Larger context than … of all models | 38% | n/a |
| Max output per response | 4K tokens (~3,100 words) | n/a |
| Input types | text, image | text |
| Output types | text | text |
| Free variant | No | No |
| Listed since | 2024-03-13 | n/a |
Monthly cost at list price for five common workloads. Token counts per unit are stated so you can scale them to your own traffic. A workload that does not fit a model's context window is marked rather than priced.
| Workload (per month) | Anthropic: Claude 3 Haiku | unsloth/Qwen3.8-27B-NVFP4 | Difference |
|---|---|---|---|
| Customer-support chatbot 50,000 conversations — each a 6-turn conversation (1,500 in / 400 out tokens) | $43.75 | n/a | — |
| RAG search over documents 20,000 questions — each a question answered from ~8 retrieved passages (6,000 in / 500 out tokens) | $42.50 | n/a | — |
| Coding / tool-using agent 2,000 tasks — each a multi-step task with tool calls and file context (60,000 in / 6,000 out tokens) | $45.00 | n/a | — |
| Batch summarisation 10,000 documents — each a 6,000-word report summarised to a page (8,000 in / 600 out tokens) | $27.50 | n/a | — |
| Long-document analysis 500 documents — each a 150-page contract or codebase read in one pass (100,000 in / 2,000 out tokens) | $13.75 | n/a | — |
| Capability | Anthropic: Claude 3 Haiku | unsloth/Qwen3.8-27B-NVFP4 | What it lets you do |
|---|---|---|---|
| Tool / function calling | ✓ Yes | — | call your APIs and run agent loops |
| Structured output (JSON schema) | — | — | return JSON that validates against your schema |
| Reasoning / thinking mode | — | — | spend extra tokens thinking before answering hard problems |
| Image input | ✓ Yes | — | read screenshots, charts and scanned pages |
| Audio input | — | — | take speech or audio directly |
| File / PDF input | — | — | accept documents without your own parsing step |
| Image output | — | — | generate images, not just text |
| Built-in web search | ✓ Yes | — | answer from live web results |
| Seed (reproducible sampling) | — | — | repeat a generation for tests and evals |
| Log-probabilities | — | — | read token confidence for classification and scoring |
From each model's published parameters and input/output types. “—” means not advertised, not proven absent.
No advantage on the data we track — pick it for provider preference or output quality on your own prompts.
Anthropic: Claude 3 Haiku comes out ahead on the data we track: 3 advantages and none for unsloth/Qwen3.8-27B-NVFP4.
Tool calling — Anthropic: Claude 3 Haiku: yes, unsloth/Qwen3.8-27B-NVFP4: not advertised. JSON-schema structured output — Anthropic: Claude 3 Haiku: not advertised, unsloth/Qwen3.8-27B-NVFP4: not advertised.
Only Anthropic: Claude 3 Haiku accepts image input; unsloth/Qwen3.8-27B-NVFP4 is text-only for input.
No free variant is listed for either; both are pay-per-token.
Claude 3 Haiku is Anthropic's fastest and most compact model for near-instant responsiveness. Quick and accurate targeted performance. See the launch announcement and benchmark results [here](https://www.anthropic.com/news/claude-3-haiku) #multimodal
Full Anthropic: Claude 3 Haiku specs →No description provided yet.
Full unsloth/Qwen3.8-27B-NVFP4 specs →