Model Release Radar
Price/capability frontier position, benchmark scores, and community signal
Model Release Radar
Claude Opus 4.8
anthropic·closed weights·Proprietary·released Thursday, May 28, 2026
Frontier position
- Behind frontier DeepSWE pass@1 (agentic coding) - GLM-5.3, GPT-5.6 Terra, Muse Spark 1.2, GPT-5.6 Sol, Grok 4.6 are priced the same or lower and at least as capable.
Cost vs. capability
Loading chart…
Benchmark scores
Measured cost per task (DeepSWE)
| Benchmark | Pass@1 | Measured cost / task | Frontier |
|---|---|---|---|
| DeepSWE (agentic coding) (max effort, 95% CI 57.2%-60.7%, 4 runs) | 59.0% | $13.22 (median $12.35) | behind |
Artificial Analysis scores (no matched cost)
Artificial Analysis scores below have no measured per-task cost, so no frontier claim is made for them here.
| Benchmark | Score |
|---|---|
| AA intelligence index | 57.3 |
| AA coding index | 74.3 |
| GPQA Diamond (science) | 92.0% |
| Humanity's Last Exam | 48.7% |
| IFBench (instruction following) | 62.2% |
| LCR | 73.0% |
| SciCode (coding) | 53.5% |
| Tau2 (tool use) | 94.4% |
| Tau2 (banking) | 34.2% |
| Terminal-Bench (hard) | 58.3% |
| Terminal-Bench 2.1 (agentic) | 84.6% |
Variants
This page covers 2 reasoning-effort variants of the same model.
| Variant | AA coding index | AA intelligence index | Price /1M |
|---|---|---|---|
| Standard (shown above) | 74.3 | 57.3 | $10/1M |
| High effort | - | - | undisclosed |
Community signal
| Variant | Coding Elo | Overall Elo | Votes |
|---|---|---|---|
| High effort | 1,490 | 1,461 | 46,773 |
| Standard | 1,485 | 1,452 | 47,459 |