The opinions, the patterns, and the posts behind them.
13 models tracked6 vendors · Source: X
97 classified mentions · 6 excludedNewest post Oct 4, 2026
THE BIG PICTURE
How the conversation is changing
PositiveNegative
36%positive (35)
11%negative (11)
49 neutral · 2 mixedNo comparable previous period
Share of classified mentions · UTCCollected opinions, unweighted by likes or reposts
READ THE EVIDENCE
In their own words
positiveJev confidence 98.0%
@fayequepeerzade Yess, actually a set up I tested and I have been really surprised by the speed and efficiency is Opus 5.5 medium/high as orchestrator, DeepSeek v4.1 flash medium as worker and DeepSeek v4.1 flash high/max as reviewer. (1/2)
Jev classification details
Overall: Positive98.0% 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
99.0%
Negative
0.0%
Neutral
1.0%
Mixed
0.0%
Not discussed
0.0%
Cannot determine
0.0%
Jev’s category decisions
Dimension
Sentiment
Confidence
Overall
Positive
98.0%
Reasoning
Not discussed
94.0%
Speed
Positive
56.0%
Cost
Not discussed
100.0%
Coding
Not discussed
81.0%
Refers to DeepSeek V4.1 FlashYes · 98.0% confidence
jev-sentiment-v4/jev-1.13.0Classified Oct 5, 2026 · UTC
neutralJev confidence 78.0%
@alfe_dc_ Right!
Fair enough. I might try this out next!
Create tasks and then very minute subtasks for each tasks.
Let's see how that goes.
Might use DeepSeek v4.1 flash for this!
Jev classification details
Overall: Neutral78.0% 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
15.0%
Negative
0.0%
Neutral
83.0%
Mixed
0.0%
Not discussed
1.0%
Cannot determine
1.0%
Jev’s category decisions
Dimension
Sentiment
Confidence
Overall
Neutral
78.0%
Reasoning
Not discussed
84.0%
Speed
Not discussed
100.0%
Cost
Not discussed
100.0%
Coding
Not discussed
99.0%
Refers to DeepSeek V4.1 FlashYes · 96.0% confidence
jev-sentiment-v4/jev-1.13.0Classified Oct 5, 2026 · UTC
negativeJev confidence 99.0%
@ArtificialAnlys I highly suggest that the weightage of https://t.co/PikJbpGPUW Benchmark should be increased Significantly !! It really shows TRUE Colors in case of many models.. like deepseek v4.1 flash which lies to your face.
Jev classification details
Overall: Negative99.0% 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
100.0%
Neutral
0.0%
Mixed
0.0%
Not discussed
0.0%
Cannot determine
0.0%
Jev’s category decisions
Dimension
Sentiment
Confidence
Overall
Negative
99.0%
Reasoning
Not discussed
83.0%
Speed
Not discussed
100.0%
Cost
Not discussed
100.0%
Coding
Not discussed
98.0%
Refers to DeepSeek V4.1 FlashYes · 100.0% confidence
jev-sentiment-v4/jev-1.13.0Classified Oct 5, 2026 · UTC
neutralJev confidence 78.0%
Qwen 4 27B dense vs GLM-5.3/DeepSeek V4.1 Flash/ Qwen 3.8 Flash.
If Qwen 4 27B dense only matches the last two generation jumps, it probably lands 42-44 on Artificial Analysis.
That still puts a single dense model in the same band as GLM-5.3 Flash and Qwen 3.8 Flash-Next people https://t.co/I0D29GNYuw https://t.co/LVGFyG3GKy
Jev classification details
Overall: Neutral78.0% 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
1.0%
Neutral
83.0%
Mixed
0.0%
Not discussed
2.0%
Cannot determine
2.0%
Jev’s category decisions
Dimension
Sentiment
Confidence
Overall
Neutral
78.0%
Reasoning
Not discussed
99.0%
Speed
Not discussed
99.0%
Cost
Not discussed
100.0%
Coding
Not discussed
100.0%
Refers to DeepSeek V4.1 FlashYes · 91.0% confidence
jev-sentiment-v4/jev-1.13.0Classified Oct 5, 2026 · UTC
neutralJev confidence 67.0%
- I got myself a bicycle
- Passed the 2tok/s with DeepSeek-V4.1-Flash
- Attended a wedding
- Reorganized my fathers tablet
Another day to be grateful :) https://t.co/GnqKrqXilQ
Jev classification details
Overall: Neutral67.0% 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
25.0%
Negative
0.0%
Neutral
72.0%
Mixed
0.0%
Not discussed
2.0%
Cannot determine
1.0%
Jev’s category decisions
Dimension
Sentiment
Confidence
Overall
Neutral
67.0%
Reasoning
Not discussed
95.0%
Speed
Neutral
24.0%
Cost
Not discussed
100.0%
Coding
Not discussed
99.0%
Refers to DeepSeek V4.1 FlashYes · 99.0% confidence
jev-sentiment-v4/jev-1.13.0Classified Oct 5, 2026 · UTC
positiveJev confidence 58.0%
@jungletek @elder_plinius I don't have enough memory to run past 64k tokens. So, I ever put an paid model to orchestrate this (like DeepSeek-V4.1-Flash, and it continuously put new chats to work, running).
Without this, I would need do manually, like 10 times, to do a task like reverse engineering...
Jev classification details
Overall: Positive58.0% 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
64.0%
Negative
2.0%
Neutral
19.0%
Mixed
12.0%
Not discussed
1.0%
Cannot determine
2.0%
Jev’s category decisions
Dimension
Sentiment
Confidence
Overall
Positive
58.0%
Reasoning
Not discussed
95.0%
Speed
Not discussed
100.0%
Cost
Not discussed
88.0%
Coding
Not discussed
61.0%
Refers to DeepSeek V4.1 FlashYes · 98.0% confidence
jev-sentiment-v4/jev-1.13.0Classified Oct 5, 2026 · UTC
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
89.0%
Neutral
7.0%
Mixed
1.0%
Not discussed
2.0%
Cannot determine
1.0%
Jev’s category decisions
Dimension
Sentiment
Confidence
Overall
Negative
86.0%
Reasoning
Not discussed
100.0%
Speed
Not discussed
100.0%
Cost
Not discussed
85.0%
Coding
Not discussed
100.0%
Refers to DeepSeek V4.1 FlashYes · 100.0% confidence
jev-sentiment-v4/jev-1.13.0Classified Oct 5, 2026 · UTC
positiveJev confidence 73.0%
Apparently it’s cheaper using DeepSeek v4.1 flash via @OpenRouter during the peak hours of @deepseek_ai 🤯
Yes, even with the 5.5% up charge. https://t.co/GjaQRmsk7h
Jev classification details
Overall: Positive73.0% 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
76.0%
Negative
1.0%
Neutral
7.0%
Mixed
15.0%
Not discussed
0.0%
Cannot determine
1.0%
Jev’s category decisions
Dimension
Sentiment
Confidence
Overall
Positive
73.0%
Reasoning
Not discussed
100.0%
Speed
Not discussed
100.0%
Cost
Positive
97.0%
Coding
Not discussed
100.0%
Refers to DeepSeek V4.1 FlashYes · 99.0% confidence
jev-sentiment-v4/jev-1.13.0Classified Oct 5, 2026 · UTC
negativeJev confidence 100.0%
Running DeepSeek-V4.1-Flash on 32GB DDR4 Ram + 2x RTX 3090
2tok/s LOL
For now I am stopping the experiment to focus on something more efficient https://t.co/xuAHN62Pc6 https://t.co/UtPzqdm4Bt
Jev classification details
Overall: Negative100.0% 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
100.0%
Neutral
0.0%
Mixed
0.0%
Not discussed
0.0%
Cannot determine
0.0%
Jev’s category decisions
Dimension
Sentiment
Confidence
Overall
Negative
100.0%
Reasoning
Not discussed
100.0%
Speed
Negative
99.0%
Cost
Not discussed
100.0%
Coding
Not discussed
100.0%
Refers to DeepSeek V4.1 FlashYes · 99.0% confidence
jev-sentiment-v4/jev-1.13.0Classified Oct 5, 2026 · UTC
negativeJev confidence 78.0%
I thought DeepSeek V4.1 Flash was the best open-weight model until I tried GLM-5.3 Flash.
GLM is more expensive per token, but it often gets to a better answer using fewer of them.
Cost/token is really not the metric that matters. Cost/task is.
Jev classification details
Overall: Negative78.0% 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
5.0%
Negative
82.0%
Neutral
1.0%
Mixed
12.0%
Not discussed
0.0%
Cannot determine
0.0%
Jev’s category decisions
Dimension
Sentiment
Confidence
Overall
Negative
78.0%
Reasoning
Not discussed
45.0%
Speed
Not discussed
100.0%
Cost
Positive
33.0%
Coding
Not discussed
99.0%
Refers to DeepSeek V4.1 FlashYes · 99.0% confidence
jev-sentiment-v4/jev-1.13.0Classified Oct 5, 2026 · UTC
positiveJev confidence 99.0%
wait wtf, deepseek v4.1 flash makes incredible ascii diagrams https://t.co/SHdvalCkkv
Jev classification details
Overall: Positive99.0% 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
99.0%
Negative
0.0%
Neutral
0.0%
Mixed
0.0%
Not discussed
0.0%
Cannot determine
1.0%
Jev’s category decisions
Dimension
Sentiment
Confidence
Overall
Positive
99.0%
Reasoning
Not discussed
99.0%
Speed
Not discussed
100.0%
Cost
Not discussed
100.0%
Coding
Not discussed
100.0%
Refers to DeepSeek V4.1 FlashYes · 100.0% confidence
jev-sentiment-v4/jev-1.13.0Classified Oct 5, 2026 · UTC
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
5.0%
Negative
0.0%
Neutral
94.0%
Mixed
0.0%
Not discussed
1.0%
Cannot determine
0.0%
Jev’s category decisions
Dimension
Sentiment
Confidence
Overall
Neutral
92.0%
Reasoning
Not discussed
100.0%
Speed
Not discussed
100.0%
Cost
Not discussed
100.0%
Coding
Not discussed
100.0%
Refers to DeepSeek V4.1 FlashYes · 99.0% confidence
jev-sentiment-v4/jev-1.13.0Classified Oct 5, 2026 · UTC
Page 1 of 9 · 97 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.Target reached · 100 / 50 accepted · target reached · Last successful collection Oct 4, 2026
Last collection attempt Oct 5, 2026 · partially completedNew analyses use TypeSafe Jev