Compare / head-to-head
Claude Haiku 4.5vsDeepSeek V3.2
Claude Haiku 4.5 and DeepSeek V3.2 split their 2 shared benchmarks evenly. Claude Haiku 4.5 has the larger context window (200K tokens).
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
1 – 1
even split
Cheaper per token
—
list price, input + output
Larger context
Claude Haiku 4.5
200K tokens
Provider uptime (30d)
100% · 100%
Anthropic · DeepSeek
Claude Haiku 4.5
Anthropic · released 2025-10-15
textvision
DeepSeek V3.2
DeepSeek · released 2025-12-01
text
- Context
- 131K
- Max out
- 66K
- Input /1M
- —
- Output /1M
- —
- Cached /1M
- —
- Scores
- 8 · 7 core
Quality
Benchmark matrix
2 shared · 2 only Claude Haiku 4.5 · 5 only DeepSeek V3.2| Benchmark | Claude Haiku 4.5 | DeepSeek V3.2 | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 2 | ||||
| SWE-bench Verified | 73.3% ↗ | 73.1% ↗ | +0.2 pt | Claude Haiku 4.5 |
| GPQA Diamond | 71.2% ↗epoch | 82.4% ↗ | -11.2 pt | DeepSeek V3.2 |
| Only Claude Haiku 4.5 reports · 2 | ||||
| OTIS Mock AIME 2024-2025 (Epoch AI run) | 66.7% ↗epoch | not reported | — | — |
| SimpleQA Verified | 13.2% ↗epoch | not reported | — | — |
| Only DeepSeek V3.2 reports · 5 | ||||
| MMLU-Pro | not reported | 85% ↗ | — | — |
| AIME 2025 | not reported | 93.1% ↗ | — | — |
| AIME 2026 | not reported | 94.2% ↗matharena | — | — |
| Humanity's Last Exam (no tools) | not reported | 25.1% ↗ | — | — |
| LMArena Elo | not reported | 1424.8 ↗ | — | — |
Scores tagged "epoch" or "matharena" are independent runs, used only where the lab has not published its own; ⚠ marks rows MathArena flags as released after the competition. Higher is better on every row. Δ is Claude Haiku 4.5 minus DeepSeek V3.2 in the benchmark's own unit. "Not reported" means the lab has not published that figure; it is not a zero. ↗ opens the source.
Specs & pricing
Side by side
Official pricing pages and model cards| Spec | Claude Haiku 4.5 | DeepSeek V3.2 | Edge |
|---|---|---|---|
| Context window | 200K tokens | 131K tokens | Claude Haiku 4.5 |
| Max output | 64K tokens | 66K tokens | DeepSeek V3.2 |
| Input price / 1M | $1 ↗ | — | — |
| Output price / 1M | $5 ↗ | — | — |
| Cached input / 1M | $0.1 ↗ | — | — |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | $6 | — | — |
| Modalities | text · vision | text | Claude Haiku 4.5 |
| Released | 2025-10-15 | 2025-12-01 | — |
| Cited benchmark scores | 4 | 8 | — |
Reliability
Provider status
Last 60 days · refreshed daily on this page · live on /statusMore matchups
Claude Haiku 4.5 vs …
Models sharing the most benchmarksMore matchups
DeepSeek V3.2 vs …
Models sharing the most benchmarksBuilt by Respan
Which one wins on your data?
Public benchmarks are a starting point. Run Claude Haiku 4.5 and DeepSeek V3.2 on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.