# cobalt-seeded-rl-base…ncp5-q4v3-groot16 by Alexander Gurung Source: https://savrn.com/models/cobalt-seeded-rl-base-ramp25-stoppen-gen4k-ep2-ncp5-q4v3-groot16 Markdown alternate of the page above; the site index is https://savrn.com/llms.txt --- ## Runs On What it takes to serve cobalt-seeded-rl-base-ramp25-stoppen-gen4k-ep2-ncp5-q4v3-groot16 (4.4B parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them. | Precision | Weights | Memory needed | Cheapest setup | Per hour | Also fits | | --- | --- | --- | --- | --- | --- | | 16-bit | 8.8 GB | 10.6 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | | 8-bit | 4.4 GB | 5.3 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | | 4-bit | 2.2 GB | 2.6 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the [SAVRN Index](https://savrn.com/ai-index/pricing/gpus), read Oct 7, 2026. [cobalt-seeded-rl-base-ramp25-stoppen-gen4k-ep2-ncp5-q4v3-groot16 on every accelerator the SAVRN Index prices, at every precision](https://savrn.com/models/cobalt-seeded-rl-base-ramp25-stoppen-gen4k-ep2-ncp5-q4v3-groot16/gpus) ## Model Card An OpenRLHF GRPO reinforcement-learning checkpoint for Qwen3-4B. - Saved at global step 60 of RL run seededrlbaseramp25stoppengen4kep2ncp5q4v3groot16. - This is the best checkpoint by pass@8 so far in this run. Trained and validated on the cobalt-train ≤2/64 frontier (canonical cleaneval prompts): 1833 train / 112 held-out val problems the base model solved on at most 2 of 64 samples under the iidcanonical@64 hardness scan. Val evals sample at temperature 1.0 (matching the cleaneval frontier eval). Reward signal: binary code-correctness (1.0 if the generated program passes the problem's tests, otherwise 0.0). This checkpoint is the main revision (git branch) of the repo, with the model at… Excerpt from the card by Alexander Gurung. ## Configuration Architecture Qwen3ForCausalLM Context length (tokens) 262,144 Layers 36 Hidden size 2,560 Feed-forward size 9,728 Attention heads 32 Key/value heads 8 Head dimension 128 Vocabulary size 151,936 Model type qwen3 ## Identity and Version Repository agurung/cobalt-seeded-rl-base-ramp25-stoppen-gen4k-ep2-ncp5-q4v3-groot16 Publisher Alexander Gurung Task Text generation Modality Text Library transformers Parameters 4.4B parameters Languages Not stated by the source Revision ac2efbeb42835b7caef7cbd3f47f8599a014d975 First published 2026-09-18 Last updated 2026-09-20 ## Files and Weights 7 files, 8.8 GB in total. The weights are 1 file totalling 8.8 GB in safetensors. Weights1 file · 8.8 GB Configuration2 files · 1.8 KB Tokenizer2 files · 11.4 MB Documentation1 file · 2.6 KB Repository1 file · 1.6 KB Every file | File | Type | Size | SHA-256 | | --- | --- | --- | --- | | model.safetensors | Weights | 8.8 GB | 1c60b312db17 | | config.json | Configuration | 1.6 KB | — | | generation_config.json | Configuration | 213 B | — | | README.md | Documentation | 2.6 KB | — | | .gitattributes | Repository | 1.6 KB | — | | tokenizer.json | Tokenizer | 11.4 MB | be75606093db | | tokenizer_config.json | Tokenizer | 694 B | — | ## License and Download License Not stated by the source Access Open weights, no gate Download size 8.8 GB [Download from Alexander Gurung](https://huggingface.co/agurung/cobalt-seeded-rl-base-ramp25-stoppen-gen4k-ep2-ncp5-q4v3-groot16) Released by Alexander Gurung through its official repository on Hugging Face. ## Built From - Derived from [Qwen/Qwen3-4B-Instruct-2507](https://savrn.com/models/qwen3-4b-instruct-2507) ## Memory Requirements | Precision | Weights in memory | | --- | --- | | As published | 8.8 GB | | 16-bit | 8.8 GB | | 8-bit | 4.4 GB | | 4-bit | 2.2 GB | Weights only, from the published parameter count; the key-value cache and runtime add to this. ## Questions About cobalt-seeded-rl-base-ramp25-stoppen-gen4k-ep2-ncp5-q4v3-groot16 ### How much GPU memory does cobalt-seeded-rl-base-ramp25-stoppen-gen4k-ep2-ncp5-q4v3-groot16 need? About 10.6 GB at 16-bit and 2.6 GB at 4-bit: the weights (4.4B parameters) plus a working margin. A long context needs more. ### What is the cheapest GPU to run cobalt-seeded-rl-base-ramp25-stoppen-gen4k-ep2-ncp5-q4v3-groot16 on? At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists. ### What is cobalt-seeded-rl-base-ramp25-stoppen-gen4k-ep2-ncp5-q4v3-groot16's context length? 262,144 tokens, from the maximum position embeddings in its published configuration. ## Similar Models Model · Text generation ### [cobalt-seeded-rl-base-ramp25-stoppen-gen4k-ep2-ncp10-q4v3-groot16](https://savrn.com/models/cobalt-seeded-rl-base-ramp25-stoppen-gen4k-ep2-ncp10-q4v3-groot16) [Alexander Gurung](https://savrn.com/model-publishers/agurung) An OpenRLHF GRPO reinforcement-learning checkpoint for Qwen3-4B. - Saved at global step 16 of RL run seededrlbaseramp25stoppengen4kep2ncp10q4v3groot16. - This is the best checkpoint by pass@8 so far in this run (evaldefaultpass8 = 0.1244). Trained and validated on the cobalt-train ≤2/64 frontier (canonical cleaneval prompts): 1833 train / 112 held-out val problems the base model solved on at most 2 of 64 samples under the iidcanonical@64 hardness scan. Val evals sample at temperature 1.0 (matching the cleaneval frontier eval). Reward signal: binary code-correctness (1.0 if the generated program passes the problem's tests, otherwise 0.0). Eval metrics at this checkpoint (held-out val, 8… Open weights 4.4B parameters 262,144 tokens transformers [View model](https://savrn.com/models/cobalt-seeded-rl-base-ramp25-stoppen-gen4k-ep2-ncp10-q4v3-groot16) Model · Text generation ### [cobalt-seeded-rl-base-ramp25-stoppen-gen4k-ep2-ncp5-q4v3-iid16](https://savrn.com/models/cobalt-seeded-rl-base-ramp25-stoppen-gen4k-ep2-ncp5-q4v3-iid16) [Alexander Gurung](https://savrn.com/model-publishers/agurung) An OpenRLHF GRPO reinforcement-learning checkpoint for Qwen3-4B. - Saved at global step 56 of RL run seededrlbaseramp25stoppengen4kep2ncp5q4v3iid16. - This is the best checkpoint by pass@8 so far in this run. Trained and validated on the cobalt-train ≤2/64 frontier (canonical cleaneval prompts): 1833 train / 112 held-out val problems the base model solved on at most 2 of 64 samples under the iidcanonical@64 hardness scan. Val evals sample at temperature 1.0 (matching the cleaneval frontier eval). Reward signal: binary code-correctness (1.0 if the generated program passes the problem's tests, otherwise 0.0). This checkpoint is the main revision (git branch) of the repo, with the model at the… Open weights 4.4B parameters 262,144 tokens transformers [View model](https://savrn.com/models/cobalt-seeded-rl-base-ramp25-stoppen-gen4k-ep2-ncp5-q4v3-iid16) Model · Text generation ### [cobalt-seeded-rl-base-ramp25-stoppen-gen4k-ep2-ncp10-base-q4v3](https://savrn.com/models/cobalt-seeded-rl-base-ramp25-stoppen-gen4k-ep2-ncp10-base-q4v3) [Alexander Gurung](https://savrn.com/model-publishers/agurung) An OpenRLHF GRPO reinforcement-learning checkpoint for Qwen3-4B. - Saved at global step 40 of RL run seededrlbaseramp25stoppengen4kep2ncp10baseq4v3. - This is the best checkpoint by pass@8 so far in this run (evaldefaultpass8 = 0.0421). Trained and validated on the cobalt-train ≤2/64 frontier (canonical cleaneval prompts): 1833 train / 112 held-out val problems the base model solved on at most 2 of 64 samples under the iidcanonical@64 hardness scan. Val evals sample at temperature 1.0 (matching the cleaneval frontier eval). Reward signal: binary code-correctness (1.0 if the generated program passes the problem's tests, otherwise 0.0). Eval metrics at this checkpoint (held-out val, 8… Open weights 4.4B parameters 262,144 tokens transformers [View model](https://savrn.com/models/cobalt-seeded-rl-base-ramp25-stoppen-gen4k-ep2-ncp10-base-q4v3) Model · Text generation ### [Qwen3-4B-Instruct-2507-FP8](https://savrn.com/models/qwen3-4b-instruct-2507-fp8) [Qwen](https://savrn.com/model-publishers/qwen) We introduce the updated version of the Qwen3-4B-FP8 non-thinking mode, named Qwen3-4B-Instruct-2507-FP8, featuring the following key enhancements: - Significant improvements in general capabilities, including instruction following, logical reasoning, text comprehension, mathematics, science, coding and tool usage. - Substantial gains in long-tail knowledge coverage across multiple languages. - Markedly better alignment with user preferences in subjective and open-ended tasks, enabling more helpful responses and higher-quality text generation. - Enhanced capabilities in 256K long-context understanding. This repo contains the FP8 version of Qwen3-4B-Instruct-2507, which has the following… Open weights apache-2.0 4.4B parameters 262,144 tokens transformers [View model](https://savrn.com/models/qwen3-4b-instruct-2507-fp8) Model · Text generation ### [Fleet-Semantic-4B](https://savrn.com/models/fleet-semantic-4b) [Noumenon, Inc.](https://savrn.com/model-publishers/noumenon-labs) Fleet is a dynamically extensible Semantic Inference Language Model (SiLM) built on Qwen/Qwen3-4B-Instruct-2507. Fleet is not an external prompt router. The routing system lives inside the causal language model itself through the custom FleetForCausalLM architecture. Fleet replaces selected Qwen projections with native FleetLinear modules. The model also contains a dynamic semantic prototype bank: Routing therefore happens inside model inference, rather than through an external classifier that chooses a different model before generation. This checkpoint currently contains six specialists: - mathreasoning - medicalreasoning - legalops - intent - functioncalling - structured The base Qwen… Open weights other 4.4B parameters 262,144 tokens transformers [View model](https://savrn.com/models/fleet-semantic-4b) Model · Text generation ### [dQwen3.5-4B-Base](https://savrn.com/models/dqwen3-5-4b-base) [IFML](https://savrn.com/model-publishers/ut-ifml) A masked diffusion language model adapted from Qwen3.5-4B. The backbone is hybrid: only its attention layers are made bidirectional, and the Gated DeltaNet layers stay causal. This is a base model, with no instruction tuning. Paper: dQwen3.5: Hybrid-Attention Diffusion Language Models. Code: https://github.com/AntonXue/dQwen Needs a CUDA GPU and transformers>=5.13 (tested with torch 2.7.1+cu128, flash-linear-attention 0.5.1). generate decodes the whole canvas at once, committing positions above a confidence threshold (tau=0.9); pass blocklength=32 for left-to-right block decoding, or tau=None, stepsperblock=k for a fixed budget. The 50B-token checkpoint from the paper is… Open weights apache-2.0 4.2B parameters 262,144 tokens transformers [View model](https://savrn.com/models/dqwen3-5-4b-base) ## Alexander Gurung [All models and datasets](https://savrn.com/model-publishers/agurung) ## Versions - [ac2efbeb4283](https://savrn.com/models/cobalt-seeded-rl-base-ramp25-stoppen-gen4k-ep2-ncp5-q4v3-groot16/versions/ac2efbeb4283) · current 2026-09-20 ## Explore More - [All text generation models](https://savrn.com/models/tasks/text-generation) - [Model comparisons](https://savrn.com/models/comparisons) - [The model directory](https://savrn.com/models) - [Open model prices by host](https://savrn.com/ai-index/pricing/open-models) ## Source - Repository metadata, read 2026-09-20. - [Hugging Face record](https://huggingface.co/agurung/cobalt-seeded-rl-base-ramp25-stoppen-gen4k-ep2-ncp5-q4v3-groot16) - [How the hub is built](https://savrn.com/model-hub/methodology)