GitHub / b4rtaz/distributed-llama issues and pull requests
#234 - The test of windows 10 failed during "make dllema"
Issue -
State: open - Opened by Tuhy97 23 days ago
- 1 comment
#233 - Implement docker
Pull Request -
State: open - Opened by enigodupont 24 days ago
#232 - Adding support for hostnames on worker connection
Pull Request -
State: closed - Opened by enigodupont 25 days ago
- 4 comments
#231 - specify binding ip
Issue -
State: open - Opened by withinboredom 26 days ago
#230 - Critical error: Cannot read magic value when first tries to inference
Issue -
State: open - Opened by Tuhy97 about 1 month ago
- 3 comments
#229 - Chatting in CJK language lead to crash
Issue -
State: closed - Opened by omegacoleman about 2 months ago
- 1 comment
#228 - fix: Tokenizer utf-8 handling
Pull Request -
State: closed - Opened by omegacoleman about 2 months ago
#227 - feat: cors support.
Pull Request -
State: closed - Opened by b4rtaz about 2 months ago
#226 - fix: memory leak in Tokenizer class
Pull Request -
State: closed - Opened by omegacoleman about 2 months ago
- 1 comment
#225 - bugs found using open-webui 'Stop Thinking' button with dllama-api
Issue -
State: closed - Opened by omegacoleman about 2 months ago
- 1 comment
#224 - fix: reset EOS detector across requests
Pull Request -
State: closed - Opened by omegacoleman about 2 months ago
- 1 comment
#223 - fix: make api server ignore SIGPIPE if needed
Pull Request -
State: closed - Opened by omegacoleman about 2 months ago
- 2 comments
#222 - feat: multiple eos tokens.
Pull Request -
State: closed - Opened by b4rtaz about 2 months ago
#221 - feat: support more than 2 eos tokens
Pull Request -
State: closed - Opened by omegacoleman about 2 months ago
- 3 comments
#220 - fix: bad bound check in shiftForward_F32_F32
Pull Request -
State: closed - Opened by omegacoleman about 2 months ago
- 1 comment
#219 - Assert failure in shiftForward_F32_F32
Issue -
State: closed - Opened by omegacoleman about 2 months ago
#218 - Access to web interface API
Issue -
State: closed - Opened by DDJBR about 2 months ago
#217 - error: ‘_MM_FROUND_TO_NEAREST_INT’ was not declared in this scope
Issue -
State: open - Opened by squidKid-deluxe about 2 months ago
- 2 comments
#216 - Support for YOLO 11 OBB
Issue -
State: open - Opened by ng-druid 2 months ago
- 1 comment
#214 - Does distributed-llama currently support multimodal models?
Issue -
State: open - Opened by SherronBurtint 2 months ago
- 2 comments
#213 - build vulkan failed using mingw
Issue -
State: open - Opened by zhangddjs 3 months ago
- 4 comments
#212 - segmentation fault
Issue -
State: open - Opened by zhangddjs 3 months ago
- 4 comments
#211 - How about Model Context Protocol support ?
Issue -
State: open - Opened by zhangddjs 3 months ago
#210 - Python pinning in converter/requirements.txt breaks pip
Issue -
State: open - Opened by notfol 3 months ago
#209 - "Network is closed" error after a few interactions in multi-node setup (2× RTX 4070)
Issue -
State: closed - Opened by tsa3 3 months ago
- 4 comments
#208 - Support for Qwen3 30ba3b
Issue -
State: open - Opened by smpurkis 3 months ago
- 1 comment
#207 - Feature Request: Enable caching
Issue -
State: closed - Opened by D-i-t-gh 3 months ago
- 1 comment
#206 - Transfer weights and the tokenizer file to the root computer.
Issue -
State: closed - Opened by evcharger 3 months ago
- 1 comment
#199 - Open WebUI support
Issue -
State: closed - Opened by Sboshoff76 3 months ago
- 2 comments
#198 - Feature Request execute in parallel
Issue -
State: open - Opened by MichaelFomenko 4 months ago
- 1 comment
#197 - Error with the "make dllama"
Issue -
State: closed - Opened by rportojr 4 months ago
- 4 comments
#196 - feat: vulkan optimization.
Pull Request -
State: closed - Opened by b4rtaz 4 months ago
#195 - fix: vulkan memory type.
Pull Request -
State: closed - Opened by b4rtaz 4 months ago
#193 - Feature Request: Multiple Workers on one Machine
Issue -
State: open - Opened by MichaelFomenko 4 months ago
- 4 comments
#192 - feat: vulkan matmul optimization.
Pull Request -
State: closed - Opened by b4rtaz 4 months ago
#191 - Network is in non-blocking mode
Issue -
State: closed - Opened by LJ-Hao 4 months ago
- 1 comment
#190 - Inconsistent struct layouts can break cross-architecture usage
Issue -
State: open - Opened by antoine-sac 4 months ago
#189 - fix: converter hf now handles byte characters. Closes #188
Pull Request -
State: closed - Opened by antoine-sac 4 months ago
- 1 comment
#188 - Vocabulary containing special "byte tokens" not converted correctly
Issue -
State: closed - Opened by antoine-sac 4 months ago
#187 - PORT Number
Issue -
State: closed - Opened by greytery 5 months ago
- 1 comment
#186 - In chat mode, the LLM agent seems to keep talking to itself without stopping.
Issue -
State: closed - Opened by Marvin-BW 5 months ago
- 3 comments
#185 - Error when converting tokenizer from Mistral Large Instruct 2411: "Exception: Cannot resolve bosId or eosIds"
Issue -
State: open - Opened by philigrale 5 months ago
- 8 comments
#184 - README.md: add model/buffer-float-type limitations
Pull Request -
State: closed - Opened by lemmi 5 months ago
- 1 comment
#183 - q80 and f16 models fail with Critical error: Unsupported ...
Issue -
State: closed - Opened by lemmi 5 months ago
- 5 comments
#182 - Converting a DeepSeek-R1 model
Issue -
State: closed - Opened by D-i-t-gh 5 months ago
- 4 comments
#181 - feat: benchmark.
Pull Request -
State: closed - Opened by b4rtaz 5 months ago
#179 - Not able see Scaling performance with NuC (12th Gen) with deepseek_r1_distill_llama_8b_q40
Issue -
State: open - Opened by deepaks2 5 months ago
- 7 comments
#178 - Downloads wrong model
Issue -
State: closed - Opened by greytery 5 months ago
- 1 comment
#177 - Convert Hugging Face model with multiple eos tokens
Issue -
State: closed - Opened by IterableTrucks 5 months ago
- 1 comment
#176 - feat: vulkan.
Pull Request -
State: closed - Opened by b4rtaz 5 months ago
#175 - Branching out on model support
Issue -
State: open - Opened by pcfreak30 5 months ago
#174 - fix: nnuint.
Pull Request -
State: closed - Opened by b4rtaz 5 months ago
#173 - Update README.md
Pull Request -
State: closed - Opened by brettp 5 months ago
- 1 comment
#172 - Launch.py - Download again test
Issue -
State: closed - Opened by greytery 5 months ago
- 1 comment
#171 - feat: quantizeF32toQ80 avx2.
Pull Request -
State: closed - Opened by b4rtaz 6 months ago
#170 - dllama-api: /v1/models: return basename of the model
Pull Request -
State: closed - Opened by lemmi 6 months ago
- 2 comments
#169 - Unable to load 405b modell on 4x64GB
Issue -
State: closed - Opened by lemmi 6 months ago
- 5 comments
#168 - fix: missing init quants in api.
Pull Request -
State: closed - Opened by b4rtaz 6 months ago
#167 - dllama-api: returns garbage
Issue -
State: closed - Opened by lemmi 6 months ago
- 1 comment
#166 - fix: fixed inference getting stuck
Pull Request -
State: closed - Opened by b4rtaz 6 months ago
#163 - feat: use softmax_F32 for sampler.
Pull Request -
State: closed - Opened by b4rtaz 6 months ago
#161 - feat: support r1 distill llama.
Pull Request -
State: closed - Opened by b4rtaz 6 months ago
#160 - fix: tokenizer utf8 support
Pull Request -
State: closed - Opened by b4rtaz 6 months ago
#159 - Converter cant build on Ubuntu 24.04
Issue -
State: closed - Opened by scm2000 6 months ago
- 2 comments
#158 - Api server
Pull Request -
State: closed - Opened by myan-o 6 months ago
#157 - Only the root device's memory is being used?
Issue -
State: closed - Opened by ethical-haquer 6 months ago
- 6 comments
#156 - feat: fundamental codebase refactor
Pull Request -
State: closed - Opened by b4rtaz 6 months ago
- 13 comments
#155 - Fixed dllama-api bug with java's HttpURLConnection
Pull Request -
State: closed - Opened by jkeegan 6 months ago
- 1 comment
#154 - build error in termux
Issue -
State: closed - Opened by myan-o 6 months ago
- 2 comments
#153 - Potential problem with dllama-api sometimes not seeing http request body?
Issue -
State: closed - Opened by jkeegan 6 months ago
- 3 comments
#152 - It's slow.
Issue -
State: closed - Opened by myan-o 6 months ago
- 13 comments
#151 - Weird bug where malformed API request causes model to analyze error message
Issue -
State: open - Opened by jkeegan 6 months ago
- 2 comments
#149 - Core dumped with ReadSocketException
Issue -
State: closed - Opened by kolchanov 7 months ago
- 9 comments
#148 - how to convert gguf in your format?
Issue -
State: closed - Opened by lexasub 7 months ago
- 1 comment
#146 - Feature request: models endpoint support in dllama-api
Issue -
State: open - Opened by jkeegan 7 months ago
#145 - Work with cline
Issue -
State: open - Opened by piotreq7 7 months ago
- 1 comment
#144 - when would llama-3.3-70B will be supported?
Issue -
State: closed - Opened by lssac 8 months ago
- 1 comment
#143 - Added help/usage to apps and new Makefile targets
Pull Request -
State: closed - Opened by jkeegan 8 months ago
- 9 comments
#142 - Unexpected Ċ, Ġ and D characters with Llama 3.3 Instruct 70b
Issue -
State: closed - Opened by lemmi 8 months ago
- 4 comments
#141 - CPU pinning not ideal on SMT
Issue -
State: closed - Opened by lemmi 8 months ago
- 2 comments
#140 - terminate called after throwing an instance of 'ReadSocketException' what(): std::exception Aborted
Issue -
State: closed - Opened by bartekmalysz94us 8 months ago
- 2 comments
#139 - Assertion `kvDim % nSlices == 0' failed - src/commands.cpp #98
Issue -
State: closed - Opened by bartekmalysz94us 8 months ago
- 3 comments
#138 - [Feature Request] Decouple the Prefill and Decode Stage
Issue -
State: closed - Opened by zhengpeirong 9 months ago
- 1 comment
#137 - fix: rms avx2 bug.
Pull Request -
State: closed - Opened by b4rtaz 9 months ago
#136 - feat: mesh topology, distributed all layers.
Pull Request -
State: closed - Opened by b4rtaz 9 months ago
- 2 comments
#135 - fix: releasing memory.
Pull Request -
State: closed - Opened by b4rtaz 9 months ago
#134 - Invalid use of free in tokenizer.cpp
Issue -
State: closed - Opened by fairydreaming 9 months ago
- 1 comment
#133 - Android devices Support
Issue -
State: closed - Opened by qtyandhasee 10 months ago
- 5 comments
#132 - Mixtral-8x7B don't work
Issue -
State: closed - Opened by MichaelFomenko 10 months ago
- 2 comments
#131 - Model is not supported: llama3_2_3b_instruct_q40
Issue -
State: closed - Opened by Znbne 10 months ago
- 6 comments
#128 - Feature Request - Add Sliding Window Memory Scheduling
Issue -
State: open - Opened by githuba9f5404 10 months ago
#127 - Segmentation fault
Issue -
State: closed - Opened by YueZhan721 10 months ago
- 7 comments
#126 - Tokenizer reported as incompatible tokenizer
Issue -
State: open - Opened by elektroinformaciobiz 10 months ago
#121 - Output all are "!"
Issue -
State: closed - Opened by HysenX-LI 11 months ago
- 6 comments
#120 - Phi-3.5 support and "KeyError: 'low_freq_factor'"
Issue -
State: closed - Opened by unclemusclez 11 months ago
#119 - dllama-api & chat ui
Issue -
State: open - Opened by twuerfl 12 months ago
#118 - feat: reduction of writeMany/readMany calls.
Pull Request -
State: closed - Opened by b4rtaz 12 months ago
#117 - update readme.md.
Pull Request -
State: closed - Opened by b4rtaz 12 months ago
#116 - Add the function of non-uniform proportional distribution based on distributed-llama
Pull Request -
State: closed - Opened by fromthefox 12 months ago
- 2 comments
#115 - Support for Gemma 2?
Issue -
State: open - Opened by sdmorrey 12 months ago
- 2 comments