ollama/ollama issues and pull requests

#4464 - Support RX6600 (gfx1032) on windows (gfx override works on linux)

Issue - State: open - Opened by usmandilmeer 4 months ago - 5 comments
Labels: bug, windows

#4434 - Behind PROXY, how to set a proxy in Ollama

Issue - State: closed - Opened by mchopra80 4 months ago - 2 comments
Labels: bug

#4429 - GPU error

Issue - State: closed - Opened by A-Akhil 4 months ago - 2 comments
Labels: bug

#4427 - ollama can't run qwen:72b, error msg ""gpu VRAM usage didn't recover within timeout

Issue - State: closed - Opened by changingshow 4 months ago - 16 comments
Labels: bug

#4360 - bge-reranker-v2-m3、mxbai-rerank-large-v1 and other rerank models

Issue - State: closed - Opened by Feng-YiJing-Dao 4 months ago - 16 comments
Labels: model request

#4359 - Mistral is not using GPU, but LLama3 is utilizing GPU properly

Issue - State: closed - Opened by itinance 4 months ago - 3 comments
Labels: bug, nvidia

#4348 - noKeyError: rms_norm_eps

Issue - State: closed - Opened by czs397001 4 months ago - 1 comment
Labels: bug

#4342 - echo 0 > /proc/sys/kernel/hung_task_timeout_secs, 然后显卡hung住了，最后服务器只能重启了

Issue - State: closed - Opened by YangFW 4 months ago - 2 comments
Labels: bug

#4341 - how to import Meta-Llama-3-120B-Instruct.imatrix

Issue - State: closed - Opened by taozhiyuai 4 months ago - 4 comments
Labels: feature request

#4307 - ollama domains when behind Web (Squid) Proxy

Issue - State: closed - Opened by StL-Jim 4 months ago - 2 comments

#4296 - Feature Request: Generate embedding for images using `/api/embeddings` endpoint

Issue - State: open - Opened by Agent-E11 4 months ago - 6 comments
Labels: feature request

#4292 - remove last bits of ParseModelPath

Pull Request - State: open - Opened by mxyng 4 months ago - 1 comment

#4291 - update push to use model.Name

Pull Request - State: open - Opened by mxyng 4 months ago - 1 comment

#4290 - fix(server): prune files

Pull Request - State: open - Opened by mxyng 4 months ago

#4274 - Update command for Linux version

Issue - State: open - Opened by Maplerxyz 4 months ago - 2 comments
Labels: feature request, linux, install

#4269 - update pull handler to use model.Name

Pull Request - State: open - Opened by mxyng 4 months ago

#4257 - Support for InternVL-Chat-V1.5

Issue - State: open - Opened by wwjCMP 4 months ago - 4 comments
Labels: model request

#4243 - moondream's abnormal reply

Issue - State: closed - Opened by taozhiyuai 4 months ago - 1 comment
Labels: bug

#4198 - Improving the efficiency of using multiple GPU cards.

Issue - State: closed - Opened by zhqfdn 5 months ago - 9 comments
Labels: feature request

#4192 - feat: support registry basic auth

Pull Request - State: open - Opened by qcu266 5 months ago - 2 comments

#4185 - Failure on pull results in Error: EOF

Issue - State: closed - Opened by MarkWard0110 5 months ago - 2 comments
Labels: bug, ollama.com

#4151 - High RAM usage causes yo-yoing memory pressure on Mac, slow inference

Issue - State: closed - Opened by joliss 5 months ago - 5 comments
Labels: bug, macos, memory

#4142 - More Quants for command-r-plus Please?

Issue - State: closed - Opened by chigkim 5 months ago - 3 comments
Labels: model request

#4139 - only 1 GPU found -- regression 1.32 -> 1.33

Issue - State: closed - Opened by AlexLJordan 5 months ago - 22 comments
Labels: bug, nvidia

#4131 - Error "timed out waiting for llama runner to start: " on larger models.

Issue - State: closed - Opened by CalvesGEH 5 months ago - 37 comments
Labels: bug

#4131 - Error "timed out waiting for llama runner to start: " on larger models.

Issue - State: closed - Opened by CalvesGEH 5 months ago - 42 comments
Labels: bug

#4077 - Stop running model without removing

Issue - State: closed - Opened by nitulkukadia 5 months ago - 9 comments
Labels: feature request

#4042 - Unable to pull model

Issue - State: open - Opened by 353167931 5 months ago - 10 comments
Labels: bug, networking

#4034 - Implement downloads via torrents

Issue - State: open - Opened by f321x 5 months ago - 4 comments
Labels: feature request, registry

#3976 - Slow downloads

Issue - State: open - Opened by Alchemistqqqq 5 months ago - 7 comments
Labels: bug, feedback wanted, registry

#3967 - The parameter is incorrect

Issue - State: closed - Opened by nb001 5 months ago - 5 comments
Labels: bug, windows

#3953 - Support VLLM as a backend

Issue - State: open - Opened by kannon92 5 months ago - 3 comments
Labels: feature request

#3938 - Massive performance regression on 0.1.32

Issue - State: closed - Opened by jsa2 5 months ago - 11 comments
Labels: bug, nvidia

#3931 - Digest mismatch, file must be downloaded again

Issue - State: closed - Opened by tttt-0814 5 months ago - 7 comments
Labels: bug

#3910 - Add OpenELM

Issue - State: open - Opened by 3Samourai 5 months ago - 15 comments
Labels: model request

#3907 - Add /api/infill for fill-in-the-middle

Pull Request - State: closed - Opened by PhilKes 5 months ago - 9 comments

#3905 - The Ollama on the Mac computer does not have a Setting button.

Issue - State: closed - Opened by liubin595338764 5 months ago - 2 comments
Labels: bug

#3848 - phi3 doesn't seem to accept SYSTEM prompt

Issue - State: closed - Opened by rb81 5 months ago - 17 comments
Labels: bug

#3827 - Enable CORS for "app://obsidian.md"

Issue - State: closed - Opened by pegasusthemis 5 months ago - 7 comments
Labels: feature request

#3816 - `i/o timeout` when running `ollama pull`

Issue - State: open - Opened by pitilessj 5 months ago - 10 comments
Labels: bug, networking

#3815 - OpenSSL SSL_read: error:0A000126

Issue - State: closed - Opened by xuya227939 5 months ago - 3 comments
Labels: bug, needs more info

#3798 - Can this library utilize Intel's UHD graphics cards?

Issue - State: closed - Opened by Meshwa428 5 months ago - 2 comments
Labels: feature request, intel

#3786 - My internet is too slow to download the model

Issue - State: open - Opened by skystar7 5 months ago - 4 comments
Labels: feature request, networking

#3749 - Rerankers and Embeddings

Issue - State: open - Opened by BradKML 5 months ago - 14 comments
Labels: model request

#3749 - Rerankers and Embeddings

Issue - State: open - Opened by BradKML 5 months ago - 14 comments
Labels: model request

#3741 - Please accept slow network connections when loading models

Issue - State: closed - Opened by igorschlum 5 months ago - 1 comment
Labels: bug, networking

#3689 - Aya by Cohere - mt5-xxl arch

Issue - State: open - Opened by oliviermills 5 months ago - 3 comments
Labels: model request

#3652 - App running in docker container can't connect to Ollama container

Issue - State: closed - Opened by thomassrour 5 months ago - 5 comments
Labels: bug

#3647 - Ollama reverts to CPU on a100 docker. "error looking up CUDA GPU memory: device memory info lookup failure 0: 4

Issue - State: closed - Opened by Yaffa16 5 months ago - 3 comments
Labels: bug, nvidia, gpu

#3643 - how to change the max input token length when I run ‘’ollama run gemma:7b-instruct-v1.1-fp16‘’

Issue - State: closed - Opened by dh12306 5 months ago - 5 comments

#3632 - StableLM-2 12B

Issue - State: closed - Opened by coder543 5 months ago - 3 comments
Labels: model request

#3606 - multilingual-e5-large and multilingual-e5-base Embedding Model Support

Issue - State: open - Opened by awilhelm-projects 5 months ago - 19 comments
Labels: model request

#3606 - multilingual-e5-large and multilingual-e5-base Embedding Model Support

Issue - State: open - Opened by awilhelm-projects 5 months ago - 19 comments
Labels: model request

#3581 - MacOS Ollama not binding to 0.0.0.0

Issue - State: closed - Opened by kellerkind84 5 months ago - 22 comments
Labels: bug

#3553 - Embedding endpoint not available on windows.

Issue - State: closed - Opened by elblogbruno 5 months ago - 2 comments
Labels: bug

#3529 - Add metrics endpoint and request metrics

Pull Request - State: open - Opened by amila-ku 5 months ago - 6 comments

#3526 - Ollama fails to start in CPU only mode

Issue - State: closed - Opened by vishnu-dev 6 months ago - 10 comments
Labels: bug

#3511 - On Windows, launching ollama from the shortcut or executable by clicking causes very slow tokens generation, but launching from commandline is fast

Issue - State: open - Opened by lrq3000 6 months ago - 32 comments
Labels: bug, windows

#3504 - I can't pull any models

Issue - State: open - Opened by jsrcode 6 months ago - 32 comments
Labels: bug

#3472 - Allow more parts of model path

Pull Request - State: open - Opened by wzshiming 6 months ago - 1 comment

#3471 - Please add Qwen-audio

Issue - State: open - Opened by zimuoo 6 months ago - 5 comments
Labels: model request

#3453 - Some ollama cli instructions: specially stop

Issue - State: closed - Opened by ejgutierrez74 6 months ago - 1 comment

#3406 - Official arm64 build does not work on Jetson Nano Orin

Issue - State: closed - Opened by gab0220 6 months ago - 21 comments
Labels: bug, nvidia

#3372 - Ollama can't run models in Docker, Certificate error x509

Issue - State: closed - Opened by BumblingWizard 6 months ago - 10 comments
Labels: bug

#3369 - Pull a model on start or without requiring serve

Issue - State: open - Opened by 0x77dev 6 months ago - 4 comments
Labels: feature request

#3368 - Reranking models

Issue - State: open - Opened by YuanfengZhang 6 months ago - 5 comments

#3314 - "server stop" and "server status" commands

Issue - State: open - Opened by FilkerZero 6 months ago - 10 comments
Labels: feature request

#3287 - Allow system message to be on its own

Issue - State: closed - Opened by jackielii 6 months ago - 3 comments
Labels: feature request

#3275 - Resumable `ollama push`

Issue - State: open - Opened by sammcj 6 months ago - 1 comment
Labels: feature request, networking

#3262 - Ollama can support windows 7?

Issue - State: closed - Opened by zhaosd 6 months ago - 5 comments

#3258 - Problem with returning response with openai compatibility

Issue - State: closed - Opened by ejgutierrez74 6 months ago - 6 comments
Labels: bug

#3240 - do not prompt to move the CLI on install flow if already installed

Pull Request - State: open - Opened by BruceMacD 6 months ago - 2 comments

#3206 - 我将MiniCPM-2B-dpo-bf16-gguf.gguf模型成功的导入到了ollama中，并运行起来了。但是在推理的过程中，发现模型再说胡话，臆想比较严重。详见截图

Issue - State: closed - Opened by zhao1012 6 months ago - 13 comments

#3203 - Allow auth between the Ollama server/client

Issue - State: open - Opened by ftoppi 6 months ago - 2 comments
Labels: feature request

#3201 - Ollama is not using my GPU (Windows)

Issue - State: closed - Opened by nethriis 6 months ago - 10 comments
Labels: bug, windows, nvidia

#3200 - ConnectionError: HTTPConnectionPool(host='localhost', port=11434): Max retries exceeded with url: /api/generate/ (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x7851288df1c0>: Failed to establish a new connection: [Errno 111] Connection refused'))

Issue - State: closed - Opened by jyotiyadav94 6 months ago - 9 comments
Labels: bug

#3189 - Add support for amd Radeon 780M gfx1103 - override works

Issue - State: open - Opened by thbley 6 months ago - 12 comments
Labels: feature request, linux, windows, amd

#3168 - WIndows security: ALERT: Trojan:Script/Wacatac.B!ml detected

Issue - State: closed - Opened by jabbor 6 months ago - 11 comments

#3168 - WIndows security: ALERT: Trojan:Script/Wacatac.B!ml detected

Issue - State: closed - Opened by jabbor 6 months ago - 11 comments

#3154 - Why Ollama is so terribly slow when I set format="json"

Issue - State: closed - Opened by eliranwong 6 months ago - 14 comments
Labels: question

#3144 - add /metrics endpoint

Issue - State: open - Opened by joecryptotoo 6 months ago - 14 comments
Labels: feature request

#3113 - Integrated Intel GPU support

Issue - State: open - Opened by clvgt12 6 months ago - 12 comments
Labels: feature request, intel

#3107 - Windows Rocm: HSA_OVERRIDE_GFX_VERSION doesn´t work

Issue - State: open - Opened by Detlev1 6 months ago - 29 comments
Labels: bug, windows, amd

#3105 - Improve usability with Bash completion for Ollama on Linux

Pull Request - State: closed - Opened by aosan 6 months ago - 5 comments

#3078 - Ollama is not using the 100% of RTX4000 VRAM (18 of 20GB)

Issue - State: open - Opened by nfsecurity 6 months ago - 13 comments
Labels: nvidia

#3067 - Additional package manager support

Issue - State: open - Opened by jmorganca 6 months ago - 7 comments
Labels: feature request, linux, windows, macos

#3021 - API endpoint for encoding and decoding tokens

Issue - State: closed - Opened by Hansson0728 6 months ago - 3 comments
Labels: feature request

#2994 - tune concurrency manager

Pull Request - State: closed - Opened by mxyng 7 months ago

#2971 - Error: Post "http://127.0.0.1:11434/api/chat": read tcp 127.0.0.1:49192->127.0.0.1:11434: wsarecv: An existing connection was forcibly closed by the remote host.

Issue - State: closed - Opened by yxl23 7 months ago - 4 comments

GitHub / ollama/ollama issues and pull requests