Benchmark scores
Higher is better for all benchmark scores (SWE-bench, GPQA, HLE, Terminal-Bench, etc.). Green marks the model with the higher score.
LLM Comparison
SWE-2 vs Claude Haiku 5.5: benchmark scores, pricing & comparison.
Side-by-side SWE-2 vs Claude Haiku 5.5 comparison across SWE-bench, GPQA, HLE, Terminal-Bench, coding agent scores, token pricing, context window, and AskClash RWT. Green marks the winner on each benchmark.
Verdict: SWE-2 ranks higher (#15 vs #16); Claude Haiku 5.5 wins 3 of 4 shared benchmarks; SWE-2 costs less ($0/$0 vs $0.10/$0.50 per 1M tokens).
Rank #15 vs #16AskClash overall scores 60.7 vs 60.0.
Pricing $0/$0 vs $0.10/$0.50Input and output token prices per 1M tokens when published.
Proprietary vs ProprietaryCognition vs Anthropic.
SWE-2 vs Claude Haiku 5.5 benchmark comparison
Green cells highlight the winning model for each metric. Scores are cached from the AskClash LLM leaderboard snapshot.
| Metric | SWE-2 | Claude Haiku 5.5 |
|---|---|---|
| Overall Score | 60.7 | 60.0 |
| Leaderboard Rank | #15 | #16 |
| ACB | 60.0 | 63.5 |
| RWT | 8.5 | — |
| Coding Agent Index | 51.9 | — |
| HLE | 43.5 | 57.4 |
| GPQA | 93.5 | — |
| SWE-Pro | — | 64.8 |
| SWE-Atlas | 66.1 | — |
| Terminal-Bench | 27.3 | 32.8 |
| DeepSWE | 73.0 | — |
| GDPval-AA | 1668.0 | 1620.0 |
| MCP Atlas | 84.2 | — |
| Finance Agent | 54.4 | — |
| CharXiv | 84.8 | — |
| MMMU-Pro | 81.6 | — |
| Input Price (per 1M tokens) | $0 | $0.10 |
| Output Price (per 1M tokens) | $0 | $0.50 |
| Context Window | 256K | 1M |
| Benchmarks Published | 13 | 8 |
SWE-2 vs Claude Haiku 5.5 head-to-head charts
SWE-2 leads 2 and Claude Haiku 5.5 leads 3 of 5 shared benchmarks. SWE-2 is cheaper on a defined 1M-input / 200K-output workload. Charts show only benchmarks both models publish.
SWE-2Claude Haiku 5.5
Overall
ACB
HLE
Terminal-Bench
GDPval-AA
SWE-2
Input$0
Output$0
Workload$0.00
Context256K
Claude Haiku 5.5
Input$0.10
Output$0.50
Workload$0.20
Context1M
Workload = published cost of 1M input + 200K output tokens. Open the live leaderboard for interactive compare charts.
More SWE-2 and Claude Haiku 5.5 comparisons
Explore how SWE-2 and Claude Haiku 5.5 stack up against other top-ranked LLMs.
How to read this comparison
Token pricing
Lower is better for input and output prices. Green marks the cheaper model per 1M tokens.
Coverage matters
Models with fewer disclosed benchmark cells may have inflated percentile scores. Check the benchmark cell count for context.
This comparison page is generated from the AskClash LLM leaderboard cache. Open the live leaderboard for real-time scores and interactive filtering.