Mintplex-Labs/anything-llm
1
0
Fork 0
mirror of https://github.com/Mintplex-Labs/anything-llm.git synced 2026-07-19 14:13:35 -04:00
Code Issues 328 Packages Projects Releases 10 Wiki Activity

[GH-ISSUE #2442] [BUG]: llama3.1 8B Context Size Max Tokens Ignored in Both Performance Modes #1589

New Issue
Closed
opened 2026-02-22 18:25:33 -05:00 by yindo · 6 comments
yindo commented 2026-02-22 18:25:33 -05:00
Owner
Copy Link
Copy Source

Originally created by @rurhrlaub on GitHub (Oct 8, 2024).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/2442

Originally assigned to: @timothycarambat on GitHub.

How are you running AnythingLLM?

AnythingLLM desktop app

What happened?

anythingfllm_context

When using "Base" as the "Performance Mode", the Max Tokens setting is ignored and Llama 3.1 is invoked with 8K context size. When setting Performance Mode to "Maximum", the Max Tokens settings is ignored and Llama 3.1 is invoked with 128K context size. Created a modelfile to enforce 32K context size but the result was 128K. Workspace was set to use the system defined LLM settings.

Are there known steps to reproduce?

See above

Originally created by @rurhrlaub on GitHub (Oct 8, 2024). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/2442 Originally assigned to: @timothycarambat on GitHub. ### How are you running AnythingLLM? AnythingLLM desktop app ### What happened? <img width="779" alt="anythingfllm_context" src="https://github.com/user-attachments/assets/bf3afd42-94fa-4716-a85a-24555377b17f"> When using "Base" as the "Performance Mode", the Max Tokens setting is ignored and Llama 3.1 is invoked with 8K context size. When setting Performance Mode to "Maximum", the Max Tokens settings is ignored and Llama 3.1 is invoked with 128K context size. Created a modelfile to enforce 32K context size but the result was 128K. Workspace was set to use the system defined LLM settings. ### Are there known steps to reproduce? See above
yindo closed this issue 2026-02-22 18:25:33 -05:00
yindo commented 2026-02-22 18:25:33 -05:00
Author
Owner
Copy Link
Copy Source

@rurhrlaub commented on GitHub (Oct 8, 2024):

Anything LLM v1.6.7

@rurhrlaub commented on GitHub (Oct 8, 2024): Anything LLM v1.6.7
yindo commented 2026-02-22 18:25:33 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Oct 9, 2024):

When using "Base" as the "Performance Mode", the Max Tokens setting is ignored and Llama 3.1 is invoked with 8K context size

This is normal and expected. See this from Ollama: https://github.com/ollama/ollama/issues/1005#issue-1977548240
And you can see what we did in this conversation: https://github.com/Mintplex-Labs/anything-llm/issues/1991

When setting Performance Mode to "Maximum", the Max Tokens settings is ignored and Llama 3.1 is invoked with 128K context size. Created a modelfile to enforce 32K context size but the result was 128K

Any parameters passed into the API will override whatever is in a Modelfile in Ollama:
https://github.com/Mintplex-Labs/anything-llm/pull/2014/files#diff-df0e7523cd11db44d61e29cfb54f0bdc2ace72ffcf18abeca888d299efd2d738R37-R40

So here, we would be passing in whatever value you have for Max Tokens in the UI. How do you see 128K and where are you seeing that?

@timothycarambat commented on GitHub (Oct 9, 2024): > When using "Base" as the "Performance Mode", the Max Tokens setting is ignored and Llama 3.1 is invoked with 8K context size This is normal and expected. See this from Ollama: https://github.com/ollama/ollama/issues/1005#issue-1977548240 And you can see what we did in this conversation: https://github.com/Mintplex-Labs/anything-llm/issues/1991 > When setting Performance Mode to "Maximum", the Max Tokens settings is ignored and Llama 3.1 is invoked with 128K context size. Created a modelfile to enforce 32K context size but the result was 128K Any parameters passed into the API will override whatever is in a Modelfile in Ollama: https://github.com/Mintplex-Labs/anything-llm/pull/2014/files#diff-df0e7523cd11db44d61e29cfb54f0bdc2ace72ffcf18abeca888d299efd2d738R37-R40 So here, we would be passing in whatever value you have for `Max Tokens` in the UI. How do you see 128K and where are you seeing that?
yindo commented 2026-02-22 18:25:33 -05:00
Author
Owner
Copy Link
Copy Source

@rurhrlaub commented on GitHub (Oct 9, 2024):

Looking at the --ctx-size parameter in the shell - it is always 8K or 128K, never 32K. 128K setting is too large to execute 8K always truncates the data in context for incomplete results:

814339519 42877 42374 4004 0 31 0 35095332 14060 - S 0 ?? 0:01.61 /Applications/AnythingLLM.app/Contents/Resources/ollama/llm serve
814339519 63122 42877 4004 0 31 0 40038304 5776448 - S 0 ?? 17:02.59 /var/folders/p2/4xbgs9lx7lvdxsffq0v123h0r8lndz/T/ollama4011639316/runners/cpu_avx2/ollama_llama_server --model /Users/ruhrlaub/Library/Application Support/anythingllm-desktop/storage/models/ollama/blobs/sha256-87048bcd55216712ef14c11c2c303728463207b165bf18440b9b84b07ec00f87 --ctx-size 8192 --batch-size 512 --embedding --log-disable --no-mmap --mlock --parallel 4 --port 57041

@rurhrlaub commented on GitHub (Oct 9, 2024): Looking at the **--ctx-size** parameter in the shell - it is always 8K or 128K, never 32K. 128K setting is too large to execute 8K always truncates the data in context for incomplete results: 814339519 42877 42374 4004 0 31 0 35095332 14060 - S 0 ?? 0:01.61 /Applications/AnythingLLM.app/Contents/Resources/ollama/llm serve 814339519 63122 42877 4004 0 31 0 40038304 5776448 - S 0 ?? 17:02.59 /var/folders/p2/4xbgs9lx7lvdxsffq0v123h0r8lndz/T/ollama4011639316/runners/cpu_avx2/ollama_llama_server --model /Users/ruhrlaub/Library/Application Support/anythingllm-desktop/storage/models/ollama/blobs/sha256-87048bcd55216712ef14c11c2c303728463207b165bf18440b9b84b07ec00f87 **--ctx-size 8192** --batch-size 512 --embedding --log-disable --no-mmap --mlock --parallel 4 --port 57041
yindo commented 2026-02-22 18:25:34 -05:00
Author
Owner
Copy Link
Copy Source

@rurhrlaub commented on GitHub (Oct 23, 2024):

Any progress or status on this? Blocking development of the next version of the Anything LLM content pack in our marketplace.

@rurhrlaub commented on GitHub (Oct 23, 2024): Any progress or status on this? Blocking development of the next version of the Anything LLM content pack in our marketplace.
yindo commented 2026-02-22 18:25:34 -05:00
Author
Owner
Copy Link
Copy Source

@blaineatnoeonai commented on GitHub (Dec 18, 2024):

We're having the same issue. We'd also like to run Ollama with a mid-sized context (128k is too much, 8k is too little).

@blaineatnoeonai commented on GitHub (Dec 18, 2024): We're having the same issue. We'd also like to run Ollama with a mid-sized context (128k is too much, 8k is too little).
yindo commented 2026-02-22 18:25:34 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Dec 18, 2024):

The log

87048bcd55216712ef14c11c2c303728463207b165bf18440b9b84b07ec00f87 --ctx-size 8192 --batch-size 512 --embedding --log-disable --no-mmap --mlock --parallel 4 --port 57041

is the n_ctx per slot

Look lower in the logs to see the real n_ctx to see if it is applied
ex: llama3.2:3b with 1000 token limit

llama_new_context_with_model: n_ctx_per_seq (1000) < n_ctx_train (xxxxx) -- the full capacity of the model will not be utilized
.....
llm_load_tensors:          CPU model buffer size =  1918.35 MiB
llama_new_context_with_model: n_seq_max     = 4
llama_new_context_with_model: n_ctx         = 4000 # because 1,000*4 slots - this is what ollama shows for --ctx-size
llama_new_context_with_model: n_ctx_per_seq = 1000 # your original input
llama_new_context_with_model: n_batch       = 2048
llama_new_context_with_model: n_ubatch      = 512
llama_new_context_with_model: flash_attn    = 0
llama_new_context_with_model: freq_base     = 500000.0
llama_new_context_with_model: freq_scale    = 1
@timothycarambat commented on GitHub (Dec 18, 2024): The log > 87048bcd55216712ef14c11c2c303728463207b165bf18440b9b84b07ec00f87 **--ctx-size 8192** --batch-size 512 --embedding --log-disable --no-mmap --mlock --parallel 4 --port 57041 is the `n_ctx` [**per slot**](https://github.com/ollama/ollama/blob/a72f2dce45f19ab724954aea1840ec6ee8f0b628/llama/runner/cache.go#L14) Look lower in the logs to see the real n_ctx to see if it is applied ex: llama3.2:3b with 1000 token limit ``` llama_new_context_with_model: n_ctx_per_seq (1000) < n_ctx_train (xxxxx) -- the full capacity of the model will not be utilized ..... llm_load_tensors: CPU model buffer size = 1918.35 MiB llama_new_context_with_model: n_seq_max = 4 llama_new_context_with_model: n_ctx = 4000 # because 1,000*4 slots - this is what ollama shows for --ctx-size llama_new_context_with_model: n_ctx_per_seq = 1000 # your original input llama_new_context_with_model: n_batch = 2048 llama_new_context_with_model: n_ubatch = 512 llama_new_context_with_model: flash_attn = 0 llama_new_context_with_model: freq_base = 500000.0 llama_new_context_with_model: freq_scale = 1 ```
yindo changed title from [BUG]: llama3.1 8B Context Size Max Tokens Ignored in Both Performance Modes to [GH-ISSUE #2442] [BUG]: llama3.1 8B Context Size Max Tokens Ignored in Both Performance Modes 2026-06-05 14:41:36 -04:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
Desktop
Docker
Integration Request
Integration Request
OS: Linux
OS: Mobile
OS: Windows
UI/UX
blocked
bug
bug
core-team-only
documentation
duplicate
embed-widget
enhancement
feature request
github_actions
good first issue
investigating
needs info / can't replicate
possible bug
pull-request
Mirrored from GitHub Pull Request
 question
stage: specifications
wontfix
No Label core-team-only investigating needs info / can't replicate possible bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Mintplex-Labs/anything-llm#1589
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?