{"work":{"id":"e73bffa4-7620-47ac-9327-259a60db52ca","openalex_id":"https://openalex.org/W4386185625","doi":"10.48550/arxiv.2308.12950","arxiv_id":"2308.12950","raw_key":null,"title":"Code Llama: Open Foundation Models for Code","authors":null,"authors_text":"Baptiste Rozi\\`ere, Jonas Gehring, Fabian Gloeckle, Sten Sootla, Itai Gat, Xiaoqing Ellen Tan","year":2023,"venue":"cs.CL","abstract":"We release Code Llama, a family of large language models for code based on Llama 2 providing state-of-the-art performance among open models, infilling capabilities, support for large input contexts, and zero-shot instruction following ability for programming tasks. We provide multiple flavors to cover a wide range of applications: foundation models (Code Llama), Python specializations (Code Llama - Python), and instruction-following models (Code Llama - Instruct) with 7B, 13B, 34B and 70B parameters each. All models are trained on sequences of 16k tokens and show improvements on inputs with up to 100k tokens. 7B, 13B and 70B Code Llama and Code Llama - Instruct variants support infilling based on surrounding content. Code Llama reaches state-of-the-art performance among open models on several code benchmarks, with scores of up to 67% and 65% on HumanEval and MBPP, respectively. Notably, Code Llama - Python 7B outperforms Llama 2 70B on HumanEval and MBPP, and all our models outperform every other publicly available model on MultiPL-E. We release Code Llama under a permissive license that allows for both research and commercial use.","external_url":"https://arxiv.org/abs/2308.12950","cited_by_count":398,"metadata_source":"pith","metadata_fetched_at":"2026-08-05T02:28:24.338817+00:00","pith_arxiv_id":"2308.12950","created_at":"2026-05-09T04:25:28.698608+00:00","updated_at":"2026-08-05T02:28:24.338817+00:00","title_quality_ok":true,"display_title":"Code Llama: Open Foundation Models for Code","render_title":"Code Llama: Open Foundation Models for Code"},"hub":{"state":{"work_id":"e73bffa4-7620-47ac-9327-259a60db52ca","tier":"super_hub","tier_reason":"100+ Pith inbound or 10,000+ external citations","pith_inbound_count":316,"external_cited_by_count":398,"distinct_field_count":18,"first_pith_cited_at":"2023-03-31T17:28:46+00:00","last_pith_cited_at":"2026-07-09T10:29:15+00:00","author_build_status":"needed","summary_status":"needed","contexts_status":"needed","graph_status":"needed","ask_index_status":"needed","reader_status":"not_needed","recognition_status":"not_needed","updated_at":"2026-08-23T16:19:29.633870+00:00","tier_text":"super_hub"},"tier":"super_hub","role_counts":[{"context_role":"background","n":39},{"context_role":"method","n":4},{"context_role":"baseline","n":3},{"context_role":"dataset","n":2},{"context_role":"other","n":1}],"polarity_counts":[{"context_polarity":"background","n":39},{"context_polarity":"use_method","n":4},{"context_polarity":"baseline","n":3},{"context_polarity":"use_dataset","n":2},{"context_polarity":"unclear","n":1}],"runs":{"ask_index":{"job_type":"ask_index","status":"succeeded","result":{"title":"Code Llama: Open Foundation Models for Code","claims":[{"claim_text":"We release Code Llama, a family of large language models for code based on Llama 2 providing state-of-the-art performance among open models, infilling capabilities, support for large input contexts, and zero-shot instruction following ability for programming tasks. We provide multiple flavors to cover a wide range of applications: foundation models (Code Llama), Python specializations (Code Llama - Python), and instruction-following models (Code Llama - Instruct) with 7B, 13B, 34B and 70B parameters each. All models are trained on sequences of 16k tokens and show improvements on inputs with up","claim_type":"abstract","evidence_strength":"source_metadata"}],"why_cited":"Pith tracks Code Llama: Open Foundation Models for Code because it crossed a citation-hub threshold.","role_counts":[]},"error":null,"updated_at":"2026-05-14T01:34:05.866031+00:00"},"author_expand":{"job_type":"author_expand","status":"succeeded","result":{"authors_linked":[{"id":"b00ff692-ff2b-4e59-b597-32001fcc9c67","orcid":null,"display_name":"Baptiste Rozi\\`ere"},{"id":"223487df-ae43-472c-8488-188ee7dd51ed","orcid":null,"display_name":"Jonas Gehring"},{"id":"03e8d0b4-151a-4810-b206-ae056aba7b82","orcid":null,"display_name":"Fabian Gloeckle"},{"id":"6ac4d159-6f74-46da-9802-8d4a5018567f","orcid":null,"display_name":"Sten Sootla"},{"id":"44434c4a-ad6b-46f6-b02d-555c5260eb9d","orcid":null,"display_name":"Itai Gat"},{"id":"89fa6f49-3d86-4fd5-bb20-ab77a335eb43","orcid":null,"display_name":"Xiaoqing Ellen Tan"}]},"error":null,"updated_at":"2026-05-14T01:34:11.861162+00:00"},"context_extract":{"job_type":"context_extract","status":"succeeded","result":{"enqueued_papers":25},"error":null,"updated_at":"2026-05-14T01:24:01.918227+00:00"},"graph_features":{"job_type":"graph_features","status":"succeeded","result":{"co_cited":[{"title":"Evaluating Large Language Models Trained on Code","work_id":"042493e9-b26f-4b4e-bbde-382072ca9b08","shared_citers":64},{"title":"DeepSeek-Coder: When the Large Language Model Meets Programming -- The Rise of Code Intelligence","work_id":"f22dae5a-27e2-41d0-a061-c4286418dee3","shared_citers":35},{"title":"Qwen2.5-Coder Technical Report","work_id":"09ba463d-6377-4017-9801-444ffb94b056","shared_citers":33},{"title":"Program Synthesis with Large Language Models","work_id":"fd241a05-03b9-4de2-9588-9d77ce176125","shared_citers":32},{"title":"GPT-4 Technical Report","work_id":"b928e041-6991-4c08-8c81-0359e4097c7b","shared_citers":29},{"title":"Llama 2: Open Foundation and Fine-Tuned Chat Models","work_id":"68a5177f-d644-44c1-bd4f-4e5278c22f5d","shared_citers":28},{"title":"StarCoder: may the source be with you!","work_id":"7e9c3d6e-d6f7-4763-9ef6-de471506c58f","shared_citers":23},{"title":"The Llama 3 Herd of Models","work_id":"1549a635-88af-4ac1-acfe-51ae7bb53345","shared_citers":23},{"title":"Qwen3 Technical Report","work_id":"25a4e30c-1232-48e7-9925-02fa12ba7c9e","shared_citers":20},{"title":"Training Verifiers to Solve Math Word Problems","work_id":"acab1aa8-b4d6-40e0-a3ee-25341701dca2","shared_citers":18},{"title":"StarCoder 2 and The Stack v2: The Next Generation","work_id":"2495cc72-f326-4c23-8a39-d9a08cf583e4","shared_citers":16},{"title":"Mistral 7B","work_id":"eb5e1305-ad11-4875-ad8d-ad8b8f697599","shared_citers":14},{"title":"Qwen Technical Report","work_id":"bb1fd52f-6b2f-437c-9516-37bdf6eb9be8","shared_citers":14},{"title":"Proximal Policy Optimization Algorithms","work_id":"240c67fe-d14d-4520-91c1-38a4e272ca19","shared_citers":13},{"title":"CodeGen: An Open Large Language Model for Code with Multi-Turn Program Synthesis","work_id":"892a192d-205a-4079-a807-00d7874b392e","shared_citers":12},{"title":"DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models","work_id":"c5006563-f3ec-438a-9e35-b7b484f34828","shared_citers":12},{"title":"DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning","work_id":"e6b75ad5-2877-4168-97c8-710407094d20","shared_citers":12},{"title":"DeepSeek-V3 Technical Report","work_id":"57d2791d-2219-4c31-a077-afc04b12a75c","shared_citers":12},{"title":"LLaMA: Open and Efficient Foundation Language Models","work_id":"c018fc23-6f3f-4035-9d02-28a2173b2b9d","shared_citers":12},{"title":"Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge","work_id":"28ea1282-d657-4c61-a83c-f1249be6d6b1","shared_citers":12},{"title":"LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code","work_id":"ea9e51ce-1e75-4182-92d8-4d25f70d2ee4","shared_citers":11},{"title":"Scaling Laws for Neural Language Models","work_id":"b7dd8749-9c45-4977-ab9b-64478dce1ae8","shared_citers":11},{"title":"Measuring Mathematical Problem Solving With the MATH Dataset","work_id":"50652ac6-fb7c-4675-a2c2-159c241feb17","shared_citers":10},{"title":"Qwen2.5 Technical Report","work_id":"d8432992-4980-4a81-85c7-9fa2c2b87f85","shared_citers":10}],"time_series":[{"n":4,"year":2023},{"n":12,"year":2024},{"n":3,"year":2025},{"n":97,"year":2026}],"dependency_candidates":[]},"error":null,"updated_at":"2026-05-14T01:34:01.561786+00:00"},"identity_refresh":{"job_type":"identity_refresh","status":"succeeded","result":{"items":[{"title":"Qwen3 Technical Report","outcome":"unchanged","work_id":"25a4e30c-1232-48e7-9925-02fa12ba7c9e","resolver":"local_arxiv","confidence":0.98,"old_work_id":"25a4e30c-1232-48e7-9925-02fa12ba7c9e"}],"counts":{"fixed":0,"merged":0,"unchanged":1,"quarantined":0,"needs_external_resolution":0},"errors":[],"attempted":1},"error":null,"updated_at":"2026-05-14T01:24:17.883291+00:00"},"role_polarity":{"job_type":"role_polarity","status":"succeeded","result":{"title":"Code Llama: Open Foundation Models for Code","claims":[{"claim_text":"We release Code Llama, a family of large language models for code based on Llama 2 providing state-of-the-art performance among open models, infilling capabilities, support for large input contexts, and zero-shot instruction following ability for programming tasks. We provide multiple flavors to cover a wide range of applications: foundation models (Code Llama), Python specializations (Code Llama - Python), and instruction-following models (Code Llama - Instruct) with 7B, 13B, 34B and 70B parameters each. All models are trained on sequences of 16k tokens and show improvements on inputs with up","claim_type":"abstract","evidence_strength":"source_metadata"}],"why_cited":"Pith tracks Code Llama: Open Foundation Models for Code because it crossed a citation-hub threshold.","role_counts":[]},"error":null,"updated_at":"2026-05-14T01:34:07.026997+00:00"},"summary_claims":{"job_type":"summary_claims","status":"succeeded","result":{"title":"Code Llama: Open Foundation Models for Code","claims":[{"claim_text":"We release Code Llama, a family of large language models for code based on Llama 2 providing state-of-the-art performance among open models, infilling capabilities, support for large input contexts, and zero-shot instruction following ability for programming tasks. We provide multiple flavors to cover a wide range of applications: foundation models (Code Llama), Python specializations (Code Llama - Python), and instruction-following models (Code Llama - Instruct) with 7B, 13B, 34B and 70B parameters each. All models are trained on sequences of 16k tokens and show improvements on inputs with up","claim_type":"abstract","evidence_strength":"source_metadata"}],"why_cited":"Pith tracks Code Llama: Open Foundation Models for Code because it crossed a citation-hub threshold.","role_counts":[]},"error":null,"updated_at":"2026-05-14T01:34:01.570596+00:00"}},"summary":{"title":"Code Llama: Open Foundation Models for Code","claims":[{"claim_text":"We release Code Llama, a family of large language models for code based on Llama 2 providing state-of-the-art performance among open models, infilling capabilities, support for large input contexts, and zero-shot instruction following ability for programming tasks. We provide multiple flavors to cover a wide range of applications: foundation models (Code Llama), Python specializations (Code Llama - Python), and instruction-following models (Code Llama - Instruct) with 7B, 13B, 34B and 70B parameters each. All models are trained on sequences of 16k tokens and show improvements on inputs with up","claim_type":"abstract","evidence_strength":"source_metadata"}],"why_cited":"Pith tracks Code Llama: Open Foundation Models for Code because it crossed a citation-hub threshold.","role_counts":[]},"graph":{"co_cited":[{"title":"Evaluating Large Language Models Trained on Code","work_id":"042493e9-b26f-4b4e-bbde-382072ca9b08","shared_citers":64},{"title":"DeepSeek-Coder: When the Large Language Model Meets Programming -- The Rise of Code Intelligence","work_id":"f22dae5a-27e2-41d0-a061-c4286418dee3","shared_citers":35},{"title":"Qwen2.5-Coder Technical Report","work_id":"09ba463d-6377-4017-9801-444ffb94b056","shared_citers":33},{"title":"Program Synthesis with Large Language Models","work_id":"fd241a05-03b9-4de2-9588-9d77ce176125","shared_citers":32},{"title":"GPT-4 Technical Report","work_id":"b928e041-6991-4c08-8c81-0359e4097c7b","shared_citers":29},{"title":"Llama 2: Open Foundation and Fine-Tuned Chat Models","work_id":"68a5177f-d644-44c1-bd4f-4e5278c22f5d","shared_citers":28},{"title":"StarCoder: may the source be with you!","work_id":"7e9c3d6e-d6f7-4763-9ef6-de471506c58f","shared_citers":23},{"title":"The Llama 3 Herd of Models","work_id":"1549a635-88af-4ac1-acfe-51ae7bb53345","shared_citers":23},{"title":"Qwen3 Technical Report","work_id":"25a4e30c-1232-48e7-9925-02fa12ba7c9e","shared_citers":20},{"title":"Training Verifiers to Solve Math Word Problems","work_id":"acab1aa8-b4d6-40e0-a3ee-25341701dca2","shared_citers":18},{"title":"StarCoder 2 and The Stack v2: The Next Generation","work_id":"2495cc72-f326-4c23-8a39-d9a08cf583e4","shared_citers":16},{"title":"Mistral 7B","work_id":"eb5e1305-ad11-4875-ad8d-ad8b8f697599","shared_citers":14},{"title":"Qwen Technical Report","work_id":"bb1fd52f-6b2f-437c-9516-37bdf6eb9be8","shared_citers":14},{"title":"Proximal Policy Optimization Algorithms","work_id":"240c67fe-d14d-4520-91c1-38a4e272ca19","shared_citers":13},{"title":"CodeGen: An Open Large Language Model for Code with Multi-Turn Program Synthesis","work_id":"892a192d-205a-4079-a807-00d7874b392e","shared_citers":12},{"title":"DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models","work_id":"c5006563-f3ec-438a-9e35-b7b484f34828","shared_citers":12},{"title":"DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning","work_id":"e6b75ad5-2877-4168-97c8-710407094d20","shared_citers":12},{"title":"DeepSeek-V3 Technical Report","work_id":"57d2791d-2219-4c31-a077-afc04b12a75c","shared_citers":12},{"title":"LLaMA: Open and Efficient Foundation Language Models","work_id":"c018fc23-6f3f-4035-9d02-28a2173b2b9d","shared_citers":12},{"title":"Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge","work_id":"28ea1282-d657-4c61-a83c-f1249be6d6b1","shared_citers":12},{"title":"LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code","work_id":"ea9e51ce-1e75-4182-92d8-4d25f70d2ee4","shared_citers":11},{"title":"Scaling Laws for Neural Language Models","work_id":"b7dd8749-9c45-4977-ab9b-64478dce1ae8","shared_citers":11},{"title":"Measuring Mathematical Problem Solving With the MATH Dataset","work_id":"50652ac6-fb7c-4675-a2c2-159c241feb17","shared_citers":10},{"title":"Qwen2.5 Technical Report","work_id":"d8432992-4980-4a81-85c7-9fa2c2b87f85","shared_citers":10}],"time_series":[{"n":4,"year":2023},{"n":12,"year":2024},{"n":3,"year":2025},{"n":97,"year":2026}],"dependency_candidates":[]},"authors":[{"id":"b00ff692-ff2b-4e59-b597-32001fcc9c67","orcid":null,"display_name":"Baptiste Rozi\\`ere","source":"manual","import_confidence":0.72},{"id":"03e8d0b4-151a-4810-b206-ae056aba7b82","orcid":null,"display_name":"Fabian Gloeckle","source":"manual","import_confidence":0.72},{"id":"44434c4a-ad6b-46f6-b02d-555c5260eb9d","orcid":null,"display_name":"Itai Gat","source":"manual","import_confidence":0.72},{"id":"223487df-ae43-472c-8488-188ee7dd51ed","orcid":null,"display_name":"Jonas Gehring","source":"manual","import_confidence":0.72},{"id":"6ac4d159-6f74-46da-9802-8d4a5018567f","orcid":null,"display_name":"Sten Sootla","source":"manual","import_confidence":0.72},{"id":"89fa6f49-3d86-4fd5-bb20-ab77a335eb43","orcid":null,"display_name":"Xiaoqing Ellen Tan","source":"manual","import_confidence":0.72}]}}