Unverified50% confidenceEventExact time
In the FlappyBit Extreme game generation task, the model thought for about 91000 tokens, inference speed dropped to 15 tok/s, and it finally completed the output at 105000 tokens after about 2 hours.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/15/2026
First Seen
Valid until: 10/29/2026
Sources
GLM 5.3 Flash Local Test: Fitting a 320B Model into 4 RTX 3090s with 1-bit Quantization
bilibili四月的尾巴_10/13/2026
Related Entities
Related Claims
UnverifiedIn the task of generating a web-based rover game, the model generated approximately 2,800 tokens of complete code in a single pass at an average output speed of 72 tokens/s, and the code ran correctly immediately after copying.75% similarUnverifiedThe author considers a model usable only if it achieves at least 10 tokens per second.71% similarUnverifiedA model running at 0.5 tok/s takes approximately 6–7 minutes to generate a 200-token code snippet.70% similarUnverifiedThe solution achieves a stable rate of approximately 40 tok/s on prose/text generation tasks and approximately 75 tok/s on programming tasks, with speed sustained throughout the entire 1M context without significant degradation.68% similarUnverifiedOnce generation begins, generating 2000 output tokens at 40 tokens per second requires an additional 50 seconds.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1036002API
curl https://kongchang.com/api/v1/knowledge/claims/1036002MCP
get_claim(id=1036002)