Navigate
Advertise here

GPT-6.1 Sol (high) vs GPT-6 Astra (low)

GPT-6.1 Sol (high) leads on average score with 9.6 vs 9.5. GPT-6.1 Sol (high) has the lower benchmark cost at $0.542 vs $1.289. GPT-6 Astra (low) is faster at 5.09s vs 13.68s, with pass rates of 93.9% vs 90.9%.

Last updated at: 2026-09-29

Compared models

Rank
#12
Total Output Tokens
38,258
Response Time (avg)
13.68s
Total Cost
$0.542
Rank
#14
Total Output Tokens
10,144
Response Time (avg)
5.09s
Total Cost
$1.289
Recommended model GPT-6.1 Sol (high)

It has the best score here (9.6), while costing about 2.4x less than GPT-6 Astra (low).

Detailed comparison

Metric GPT-6.1 Sol GPT-6.1 Sol high Release: 2026-09-29 GPT-6 Astra GPT-6 Astra low Release: 2026-09-04
Score 9.6 9.5
Rank #12 #14
Reliability 10.0 9.9
Consistency 9.6 10.0
Attempts 66/66 66/66
Tests Correct
Attempt pass rate 93.9% 90.9%
Flaky tests 1 0
Total Runs 66 66
Cost per result 2.708 6.441
Total Cost $0.542 $1.289
Input Price $2.000 / 1M $10.000 / 1M
Output Price $10.000 / 1M $50.000 / 1M
Total Input Tokens 79,492 78,100
Output Tokens 4,954 4,155
Reasoning Tokens 33,304 5,989
Response Time (avg) 13.68s 5.09s
Response Time (max) 139.76s 28.60s
Response Time (total) 301.03s 112.04s
Parameters ~2T total (~150B active) ~2T total (~150B active)
Availability Closed Closed

Model generation showcase

Hamster playing table tennis

Prompt: Create a detailed SVG illustration of a hamster playing table tennis.

#12 GPT-6.1 Sol

high
Cost
$0.079
Time
149.5s
Tokens
7,920 tok

#14 GPT-6 Astra

low
Invalid SVG
Cost
$0.151
Time
48.9s
Tokens
3,094 tok

Top Models by Score

Score vs Total Cost

Response Time (avg)

Score vs Response Time (avg)

Total Output Tokens

Score vs Total Output Tokens

Category Breakdown

Coding Score Consistency Attempt pass rate Flaky tests Tests Correct Response Time (avg) Input Tokens Output Tokens Reasoning Tokens
GPT-6.1 Sol 10.0 10.0 100.0% 0 5.18s 7,302 390 2,215
GPT-6 Astra 10.0 10.0 100.0% 0 3.97s 7,302 388 976

Quick Compare

Switch Comparison Pair