Model Release Radar
Price/capability frontier position, benchmark scores, and community signal
Model Release Radar
GPT-6.2 Astra
openaiยทclosed weightsยทProprietaryยทreleased Tuesday, Nov 17, 2026
USD per million tokens via OpenRouter, OpenAI (openai). Checked 2026-11-20 17:14 UTC.
| Token type | USD /1M |
|---|---|
| Input | $8 |
| Output | $40 |
| Cache read | $0.4 |
| Cache write (30 minutes) | $10 |
| Cache write (1 hour) | Unknown |
Cache-write rates with a verified basis replace the ordinary input charge. Unknown rates are never treated as zero. Long-context tiers apply; use the comparison page for a prompt length estimate.
- Behind frontier AA intelligence index (blended per-1M-token price, 3:1 input/output; not a measured per-task cost) - Claude Fable 5.2, Claude Opus 5.5, Gemini 4 Neon, Gemini 4 Carbon, GPT-6.2 Sol are priced the same or lower and at least as capable.
This model has not been measured on DeepSWE (or any other benchmark with a real measured per-task cost) yet, so no cost/capability chart is drawn. Its Artificial Analysis benchmark scores are listed below.
Measured cost per task (DeepSWE)
This model has not been measured on DeepSWE (or any other benchmark with a real measured per-task cost) yet.
Artificial Analysis scores
Artificial Analysis scores describe capability. Compare them against estimated token spend with your cache mix; these estimates are not measured task costs.
| Benchmark | Score |
|---|---|
| AA intelligence index | 62.1 |
This page covers 5 reasoning-effort variants of the same model.
| Variant | AA intelligence index | Blended /1M (3:1 input/output) |
|---|---|---|
| Max effort (shown above) | 62.1 | $16/1M |
| Extra-high effort | 61.6 | $16/1M |
| High effort | 60.1 | $16/1M |
| Low effort | 45.7 | $16/1M |
| Medium effort | 57.0 | $16/1M |
| Variant | Coding Elo | Overall Elo | Votes |
|---|---|---|---|
| Max effort | 1,450 | 1,414 | 2,398 |