The opinions, the patterns, and the posts behind them.
13 models tracked6 vendors · Source: X
146 classified mentions · 8 excludedNewest post Oct 5, 2026
THE BIG PICTURE
How the conversation is changing
PositiveNegative
29%positive (42)
17%negative (25)
68 neutral · 11 mixedNo comparable previous period
Share of classified mentions · UTCCollected opinions, unweighted by likes or reposts
READ THE EVIDENCE
In their own words
neutralJev confidence 94.0%
Frontier models shipped in a cluster mid-September.
Grok 4.7, GPT-6 Sol, GPT-6 Luna, Opus 5.5.
A month later the useful question is which one still earns its seat on real jobs, not launch week.
Jev classification details
Overall: Neutral94.0% confidence
The model is discussed in this dimension without a clear positive or negative opinion.
Confidence is Jev’s estimate, not verified accuracy. These are the saved classification results; Jev does not provide a written explanation for this post.
Overall label probabilities
Positive
0.0%
Negative
0.0%
Neutral
95.0%
Mixed
0.0%
Not discussed
4.0%
Cannot determine
1.0%
Jev’s category decisions
Dimension
Sentiment
Confidence
Overall
Neutral
94.0%
Reasoning
Not discussed
99.0%
Speed
Not discussed
100.0%
Cost
Not discussed
100.0%
Coding
Not discussed
100.0%
Refers to GPT-6 LunaYes · 96.0% confidence
jev-sentiment-v4/jev-1.13.0Classified Oct 5, 2026 · UTC
negativeJev confidence 46.0%
@VictorTaelin btw been trying to set OptChat up for myself with `pi` today and I found Jev works well as a quick and cheap way to scan summaries for hallucinations and irrelevant added details.
Only seems to happen ~5% of the time (gpt-6-luna on low), but still!
also apparently the "SCALE
Jev classification details
Overall: Negative46.0% confidence
Criticism, disappointment, or an unfavorable opinion about the model in this dimension.
Confidence is Jev’s estimate, not verified accuracy. These are the saved classification results; Jev does not provide a written explanation for this post.
Overall label probabilities
Positive
2.0%
Negative
55.0%
Neutral
18.0%
Mixed
18.0%
Not discussed
1.0%
Cannot determine
6.0%
Jev’s category decisions
Dimension
Sentiment
Confidence
Overall
Negative
46.0%
Reasoning
Not discussed
60.0%
Speed
Not discussed
99.0%
Cost
Not discussed
93.0%
Coding
Not discussed
92.0%
Refers to GPT-6 LunaYes · 68.0% confidence
jev-sentiment-v4/jev-1.13.0Classified Oct 5, 2026 · UTC
neutralJev confidence 50.0%
@MLCatttt @t_ed_li @junyu_r @BanburyRoadAI @speedrun I have one very specific question, does it beat GPT 6 Luna ? If the answer is no then its dead on arrival, and its very difficult one to beat it at its price point, time per task and cost per task.
Jev classification details
Overall: Neutral50.0% confidence
The model is discussed in this dimension without a clear positive or negative opinion.
Confidence is Jev’s estimate, not verified accuracy. These are the saved classification results; Jev does not provide a written explanation for this post.
Overall label probabilities
Positive
12.0%
Negative
12.0%
Neutral
59.0%
Mixed
3.0%
Not discussed
1.0%
Cannot determine
13.0%
Jev’s category decisions
Dimension
Sentiment
Confidence
Overall
Neutral
50.0%
Reasoning
Not discussed
91.0%
Speed
Not discussed
81.0%
Cost
Cannot determine
30.0%
Coding
Not discussed
99.0%
Refers to GPT-6 LunaYes · 90.0% confidence
jev-sentiment-v4/jev-1.13.0Classified Oct 5, 2026 · UTC
neutralJev confidence 32.0%
@MLCatttt @t_ed_li @junyu_r @BanburyRoadAI @speedrun Why don't you compare your model /w e.g. GPT-6 Luna?
I bet the efficiency advantage wouldn't be as big. GPT 5.5 is outdated by multiple gens.
Comparing a new model /w an old competitor one is a tempting, but misleading comparison.
Personally, I don't trust companies that do this.
Jev classification details
Overall: Neutral32.0% confidence
The model is discussed in this dimension without a clear positive or negative opinion.
Confidence is Jev’s estimate, not verified accuracy. These are the saved classification results; Jev does not provide a written explanation for this post.
Overall label probabilities
Positive
18.0%
Negative
32.0%
Neutral
44.0%
Mixed
1.0%
Not discussed
2.0%
Cannot determine
3.0%
Jev’s category decisions
Dimension
Sentiment
Confidence
Overall
Neutral
32.0%
Reasoning
Not discussed
98.0%
Speed
Not discussed
98.0%
Cost
Not discussed
100.0%
Coding
Not discussed
100.0%
Refers to GPT-6 LunaYes · 90.0% confidence
jev-sentiment-v4/jev-1.13.0Classified Oct 5, 2026 · UTC
positiveJev confidence 89.0%
AI cost depends on more than list price: cached input, batch processing, prompt length and reasoning effort matter. Use GPT-6 Luna for volume, GPT-6.1 Sol for quality, and Astra only for the hardest tasks. Test your workload before switching. https://t.co/Pe5L5mvmYH
Jev classification details
Overall: Positive89.0% confidence
Praise, endorsement, or a favorable opinion about the model in this dimension.
Confidence is Jev’s estimate, not verified accuracy. These are the saved classification results; Jev does not provide a written explanation for this post.
Overall label probabilities
Positive
91.0%
Negative
0.0%
Neutral
6.0%
Mixed
3.0%
Not discussed
0.0%
Cannot determine
0.0%
Jev’s category decisions
Dimension
Sentiment
Confidence
Overall
Positive
89.0%
Reasoning
Not discussed
80.0%
Speed
Not discussed
100.0%
Cost
Positive
60.0%
Coding
Not discussed
100.0%
Refers to GPT-6 LunaYes · 97.0% confidence
jev-sentiment-v4/jev-1.13.0Classified Oct 5, 2026 · UTC
positiveJev confidence 66.0%
Simple model-selection rule: GPT-6 Luna for high-volume, cost-sensitive tasks; GPT-6.1 Sol for stronger coding and reasoning; GPT-6 Astra for the hardest professional workflows. Match capability to the job—not just the model number. https://t.co/8GzokB99JP
Jev classification details
Overall: Positive66.0% confidence
Praise, endorsement, or a favorable opinion about the model in this dimension.
Confidence is Jev’s estimate, not verified accuracy. These are the saved classification results; Jev does not provide a written explanation for this post.
Overall label probabilities
Positive
73.0%
Negative
0.0%
Neutral
25.0%
Mixed
2.0%
Not discussed
0.0%
Cannot determine
0.0%
Jev’s category decisions
Dimension
Sentiment
Confidence
Overall
Positive
66.0%
Reasoning
Not discussed
93.0%
Speed
Not discussed
100.0%
Cost
Positive
89.0%
Coding
Not discussed
74.0%
Refers to GPT-6 LunaYes · 99.0% confidence
jev-sentiment-v4/jev-1.13.0Classified Oct 5, 2026 · UTC
neutralJev confidence 69.0%
Quick pricing lesson for AI builders: GPT-6 Luna is currently the cheapest OpenAI reasoning model at $0.10 input / $0.50 output per 1M tokens. GPT-6.1 Sol is $2 / $10, while GPT-6 Astra is $10 / $50. Choose quality based on the task—not the model number. https://t.co/8GzokB99JP
Jev classification details
Overall: Neutral69.0% confidence
The model is discussed in this dimension without a clear positive or negative opinion.
Confidence is Jev’s estimate, not verified accuracy. These are the saved classification results; Jev does not provide a written explanation for this post.
Overall label probabilities
Positive
23.0%
Negative
1.0%
Neutral
75.0%
Mixed
1.0%
Not discussed
0.0%
Cannot determine
0.0%
Jev’s category decisions
Dimension
Sentiment
Confidence
Overall
Neutral
69.0%
Reasoning
Not discussed
95.0%
Speed
Not discussed
100.0%
Cost
Positive
98.0%
Coding
Not discussed
100.0%
Refers to GPT-6 LunaYes · 100.0% confidence
jev-sentiment-v4/jev-1.13.0Classified Oct 5, 2026 · UTC
negativeJev confidence 43.0%
@DavidSHolz GPT 6 luna and Sol are just...
Luna even said to just go back to Luna 5.6. unprompted. after I swapped and said hi.
Sol 6 only answers everything creative with two sentences. refuses to do anything that's a paragraph. I don't know what's wrong with them.
Jev classification details
Overall: Negative43.0% confidence
Criticism, disappointment, or an unfavorable opinion about the model in this dimension.
Confidence is Jev’s estimate, not verified accuracy. These are the saved classification results; Jev does not provide a written explanation for this post.
Overall label probabilities
Positive
11.0%
Negative
53.0%
Neutral
15.0%
Mixed
11.0%
Not discussed
1.0%
Cannot determine
9.0%
Jev’s category decisions
Dimension
Sentiment
Confidence
Overall
Negative
43.0%
Reasoning
Not discussed
57.0%
Speed
Not discussed
100.0%
Cost
Not discussed
100.0%
Coding
Not discussed
100.0%
Refers to GPT-6 LunaYes · 93.0% confidence
jev-sentiment-v4/jev-1.13.0Classified Oct 5, 2026 · UTC
negativeJev confidence 98.0%
@darekgusto 😭😭 he said the same thing just few mins back and if u test gpt 6 luna now that is even worse in speed right now compared to 6.1 sol
Jev classification details
Overall: Negative98.0% confidence
Criticism, disappointment, or an unfavorable opinion about the model in this dimension.
Confidence is Jev’s estimate, not verified accuracy. These are the saved classification results; Jev does not provide a written explanation for this post.
Overall label probabilities
Positive
0.0%
Negative
98.0%
Neutral
0.0%
Mixed
0.0%
Not discussed
0.0%
Cannot determine
2.0%
Jev’s category decisions
Dimension
Sentiment
Confidence
Overall
Negative
98.0%
Reasoning
Not discussed
98.0%
Speed
Negative
93.0%
Cost
Not discussed
100.0%
Coding
Not discussed
99.0%
Refers to GPT-6 LunaYes · 87.0% confidence
jev-sentiment-v4/jev-1.13.0Classified Oct 5, 2026 · UTC
mixedJev confidence 100.0%
@thsottiaux @thsottiaux Any chance we can get a speedup on GPT-6 Luna? The model is great and I love how little usage it uses up, but it's so painfully slow 😭
Jev classification details
Overall: Mixed100.0% confidence
Both benefits and drawbacks are expressed about the same model in this dimension.
Confidence is Jev’s estimate, not verified accuracy. These are the saved classification results; Jev does not provide a written explanation for this post.
Overall label probabilities
Positive
0.0%
Negative
0.0%
Neutral
0.0%
Mixed
100.0%
Not discussed
0.0%
Cannot determine
0.0%
Jev’s category decisions
Dimension
Sentiment
Confidence
Overall
Mixed
100.0%
Reasoning
Not discussed
100.0%
Speed
Negative
100.0%
Cost
Not discussed
63.0%
Coding
Not discussed
100.0%
Refers to GPT-6 LunaYes · 98.0% confidence
jev-sentiment-v4/jev-1.13.0Classified Oct 5, 2026 · UTC
positiveJev confidence 100.0%
@HeyAliux A lot of companies are moving go GPT 6 Luna for that reason. Cost. Its dirt cheap, and can do almost everything the other models can do. Anthropic not in a good shape. Supply Chain Risk and no answer to 6 Luna is bad bad bad
Jev classification details
Overall: Positive100.0% confidence
Praise, endorsement, or a favorable opinion about the model in this dimension.
Confidence is Jev’s estimate, not verified accuracy. These are the saved classification results; Jev does not provide a written explanation for this post.
Overall label probabilities
Positive
100.0%
Negative
0.0%
Neutral
0.0%
Mixed
0.0%
Not discussed
0.0%
Cannot determine
0.0%
Jev’s category decisions
Dimension
Sentiment
Confidence
Overall
Positive
100.0%
Reasoning
Not discussed
99.0%
Speed
Not discussed
100.0%
Cost
Positive
100.0%
Coding
Not discussed
100.0%
Refers to GPT-6 LunaYes · 94.0% confidence
jev-sentiment-v4/jev-1.13.0Classified Oct 5, 2026 · UTC
negativeJev confidence 96.0%
@Bhavani_00007 its not true and they made gpt 6 luna even slower with this thing
Jev classification details
Overall: Negative96.0% confidence
Criticism, disappointment, or an unfavorable opinion about the model in this dimension.
Confidence is Jev’s estimate, not verified accuracy. These are the saved classification results; Jev does not provide a written explanation for this post.
Overall label probabilities
Positive
0.0%
Negative
97.0%
Neutral
0.0%
Mixed
0.0%
Not discussed
0.0%
Cannot determine
3.0%
Jev’s category decisions
Dimension
Sentiment
Confidence
Overall
Negative
96.0%
Reasoning
Not discussed
96.0%
Speed
Negative
89.0%
Cost
Not discussed
100.0%
Coding
Not discussed
98.0%
Refers to GPT-6 LunaYes · 87.0% confidence
jev-sentiment-v4/jev-1.13.0Classified Oct 5, 2026 · UTC
Page 1 of 13 · 146 mentions
1 of 1 enabled models have a recorded collection outcome with target 50 in the last 7 days. Search exhaustion and page limits can produce fewer records.Page limit reached · 46 / 50 accepted · page cap · Last successful collection Oct 4, 2026
Last collection attempt Oct 5, 2026 · partially completedNew analyses use TypeSafe Jev