{ "id": "qwen38-flash-next-ple-memory", "title": "Qwen Flash-Next: quantized PLE and memory residence", "kind": "benchmark", "source_type": "community_operator_report", "summary": "This experiment changes the n-gram PLE table backing, rather than quantizing every model component. The source compares a BF16 table held in RAM with file-mapped sidecars on one 96 GB RTX PRO 6000 and a 176 GB host. Main-table throughput uses 8K input, 512 requested output tokens and no prefix-cache hits. Full checkpoint revision and executed image digest are not supplied.", "runtime": "vLLM", "runtime_topics": [ "vLLM" ], "hardware_topics": [ "nvidia" ], "platform_topics": [ "linux" ], "model_label": "Qwen3.8-Flash-Next", "model_family": "Qwen", "model_id": "primitive-ai/Qwen3.8-Flash-Next-mixed-NVFP4-FP8", "model_release_date": null, "model_artifact": { "id": "primitive-ai/Qwen3.8-Flash-Next-mixed-NVFP4-FP8", "revision": null }, "evidence_state": "source_reported", "site_reproduced": false, "record_type": "curated_reference", "performance_ranking_allowed": false, "metrics": null, "metrics_status": "site_metrics_unmeasured; attributed_source_figures_in_editorial.reported_results", "conditions": { "model_revision": null, "weight_hash": null, "runtime_execution_revision": null, "input_tokens": null, "output_tokens": null, "concurrency": null, "cache_state": null, "quantization": null, "speculative_decoding": null, "temperature": null, "top_p": null }, "reported_setup": { "system": "one RTX PRO 6000 Blackwell 96 GB, 176 GB host RAM, local NVMe", "node_count": 1, "parallelism": null, "kv_dtype": null, "context_limit_tokens": 32768, "experiment_date": null }, "test_profiles": [ { "label": "BF16 in RAM table size", "evidence_url": "https://huggingface.co/primitive-ai/Qwen3.8-Flash-Next-PLE-quant/blob/a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763/README.md#L96", "conditions": { "model": "Qwen3.8-Flash-Next", "model_artifact": "primitive-ai/Qwen3.8-Flash-Next-mixed-NVFP4-FP8", "model_revision": null, "PLE_artifact": "primitive-ai/Qwen3.8-Flash-Next-PLE-quant", "PLE_artifact_revision": "a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763", "system": "one RTX PRO 6000 Blackwell 96 GB, 176 GB host RAM, local NVMe", "node_count": 1, "engine": "vLLM with exact-image PLE overlay", "engine_version": null, "runtime_execution_revision": null, "image_tag": "vllm/vllm-openai:qwen38-flash-next", "executed_image_digest": null, "experiment_date": null, "source_update_boundary": "2026-09-06 corrections present; individual table trial dates not stated", "model_weight_precision": "mixed NVFP4/FP8 checkpoint; PLE format varies", "context_limit_tokens": 32768, "configured_max_requests": 36, "input_tokens": 8192, "output_tokens": 512, "actual_output_tokens": null, "concurrency": 1, "cache_state": "prefix-cache-free; mmap page-cache residency not independently logged", "thinking": true, "temperature": null, "top_p": null, "draft_tokens": null, "MTP_main_table_execution": null, "kv_dtype": null, "throughput_repetitions": "two seeds, a/b; not independent boots", "memory_sampling_instant": null, "reported_size_unit": "GB as labelled by source, byte convention not independently specified", "PLE_format": "BF16 in RAM", "memory_kind": "PLE_artifact_size", "memory_backing": "in RAM" } }, { "label": "BF16 in RAM host RSS", "evidence_url": "https://huggingface.co/primitive-ai/Qwen3.8-Flash-Next-PLE-quant/blob/a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763/README.md#L96", "conditions": { "model": "Qwen3.8-Flash-Next", "model_artifact": "primitive-ai/Qwen3.8-Flash-Next-mixed-NVFP4-FP8", "model_revision": null, "PLE_artifact": "primitive-ai/Qwen3.8-Flash-Next-PLE-quant", "PLE_artifact_revision": "a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763", "system": "one RTX PRO 6000 Blackwell 96 GB, 176 GB host RAM, local NVMe", "node_count": 1, "engine": "vLLM with exact-image PLE overlay", "engine_version": null, "runtime_execution_revision": null, "image_tag": "vllm/vllm-openai:qwen38-flash-next", "executed_image_digest": null, "experiment_date": null, "source_update_boundary": "2026-09-06 corrections present; individual table trial dates not stated", "model_weight_precision": "mixed NVFP4/FP8 checkpoint; PLE format varies", "context_limit_tokens": 32768, "configured_max_requests": 36, "input_tokens": 8192, "output_tokens": 512, "actual_output_tokens": null, "concurrency": 1, "cache_state": "prefix-cache-free; mmap page-cache residency not independently logged", "thinking": true, "temperature": null, "top_p": null, "draft_tokens": null, "MTP_main_table_execution": null, "kv_dtype": null, "throughput_repetitions": "two seeds, a/b; not independent boots", "memory_sampling_instant": null, "reported_size_unit": "GB as labelled by source, byte convention not independently specified", "PLE_format": "BF16 in RAM", "memory_kind": "host_RSS_not_necessarily_anonymous", "memory_backing": "in RAM" } }, { "label": "INT4 group-16 mmap table size", "evidence_url": "https://huggingface.co/primitive-ai/Qwen3.8-Flash-Next-PLE-quant/blob/a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763/README.md#L98", "conditions": { "model": "Qwen3.8-Flash-Next", "model_artifact": "primitive-ai/Qwen3.8-Flash-Next-mixed-NVFP4-FP8", "model_revision": null, "PLE_artifact": "primitive-ai/Qwen3.8-Flash-Next-PLE-quant", "PLE_artifact_revision": "a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763", "system": "one RTX PRO 6000 Blackwell 96 GB, 176 GB host RAM, local NVMe", "node_count": 1, "engine": "vLLM with exact-image PLE overlay", "engine_version": null, "runtime_execution_revision": null, "image_tag": "vllm/vllm-openai:qwen38-flash-next", "executed_image_digest": null, "experiment_date": null, "source_update_boundary": "2026-09-06 corrections present; individual table trial dates not stated", "model_weight_precision": "mixed NVFP4/FP8 checkpoint; PLE format varies", "context_limit_tokens": 32768, "configured_max_requests": 36, "input_tokens": 8192, "output_tokens": 512, "actual_output_tokens": null, "concurrency": 1, "cache_state": "prefix-cache-free; mmap page-cache residency not independently logged", "thinking": true, "temperature": null, "top_p": null, "draft_tokens": null, "MTP_main_table_execution": null, "kv_dtype": null, "throughput_repetitions": "two seeds, a/b; not independent boots", "memory_sampling_instant": null, "reported_size_unit": "GB as labelled by source, byte convention not independently specified", "PLE_format": "INT4 group-16 mmap", "memory_kind": "PLE_artifact_size", "memory_backing": "reclaimable mapped file pages" } }, { "label": "INT4 group-16 mmap host RSS", "evidence_url": "https://huggingface.co/primitive-ai/Qwen3.8-Flash-Next-PLE-quant/blob/a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763/README.md#L98", "conditions": { "model": "Qwen3.8-Flash-Next", "model_artifact": "primitive-ai/Qwen3.8-Flash-Next-mixed-NVFP4-FP8", "model_revision": null, "PLE_artifact": "primitive-ai/Qwen3.8-Flash-Next-PLE-quant", "PLE_artifact_revision": "a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763", "system": "one RTX PRO 6000 Blackwell 96 GB, 176 GB host RAM, local NVMe", "node_count": 1, "engine": "vLLM with exact-image PLE overlay", "engine_version": null, "runtime_execution_revision": null, "image_tag": "vllm/vllm-openai:qwen38-flash-next", "executed_image_digest": null, "experiment_date": null, "source_update_boundary": "2026-09-06 corrections present; individual table trial dates not stated", "model_weight_precision": "mixed NVFP4/FP8 checkpoint; PLE format varies", "context_limit_tokens": 32768, "configured_max_requests": 36, "input_tokens": 8192, "output_tokens": 512, "actual_output_tokens": null, "concurrency": 1, "cache_state": "prefix-cache-free; mmap page-cache residency not independently logged", "thinking": true, "temperature": null, "top_p": null, "draft_tokens": null, "MTP_main_table_execution": null, "kv_dtype": null, "throughput_repetitions": "two seeds, a/b; not independent boots", "memory_sampling_instant": null, "reported_size_unit": "GB as labelled by source, byte convention not independently specified", "PLE_format": "INT4 group-16 mmap", "memory_kind": "host_RSS_not_necessarily_anonymous", "memory_backing": "reclaimable mapped file pages" } }, { "label": "mmap cold first request", "evidence_url": "https://huggingface.co/primitive-ai/Qwen3.8-Flash-Next-PLE-quant/blob/a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763/README.md#L68", "conditions": { "model": "Qwen3.8-Flash-Next", "model_artifact": "primitive-ai/Qwen3.8-Flash-Next-mixed-NVFP4-FP8", "model_revision": null, "PLE_artifact": "primitive-ai/Qwen3.8-Flash-Next-PLE-quant", "PLE_artifact_revision": "a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763", "system": "one RTX PRO 6000 Blackwell 96 GB, 176 GB host RAM, local NVMe", "node_count": 1, "engine": "vLLM with exact-image PLE overlay", "engine_version": null, "runtime_execution_revision": null, "image_tag": "vllm/vllm-openai:qwen38-flash-next", "executed_image_digest": null, "experiment_date": null, "source_update_boundary": "2026-09-06 corrections present; individual table trial dates not stated", "model_weight_precision": "mixed NVFP4/FP8 checkpoint; PLE format varies", "context_limit_tokens": null, "configured_max_requests": null, "input_tokens": null, "output_tokens": null, "actual_output_tokens": null, "concurrency": 1, "cache_state": "page cache dropped before each arm", "thinking": null, "temperature": null, "top_p": null, "draft_tokens": 0, "MTP_main_table_execution": null, "kv_dtype": null, "throughput_repetitions": null, "memory_sampling_instant": null, "reported_size_unit": "GB as labelled by source, byte convention not independently specified", "PLE_format": "INT4 group-16", "workload": "first request, exact request shape unreported; optional resident-vs-mmap cohort, not main 8K/512 table", "memory_backing": "NVMe mmap" } }, { "label": "resident cold first request", "evidence_url": "https://huggingface.co/primitive-ai/Qwen3.8-Flash-Next-PLE-quant/blob/a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763/README.md#L69", "conditions": { "model": "Qwen3.8-Flash-Next", "model_artifact": "primitive-ai/Qwen3.8-Flash-Next-mixed-NVFP4-FP8", "model_revision": null, "PLE_artifact": "primitive-ai/Qwen3.8-Flash-Next-PLE-quant", "PLE_artifact_revision": "a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763", "system": "one RTX PRO 6000 Blackwell 96 GB, 176 GB host RAM, local NVMe", "node_count": 1, "engine": "vLLM with exact-image PLE overlay", "engine_version": null, "runtime_execution_revision": null, "image_tag": "vllm/vllm-openai:qwen38-flash-next", "executed_image_digest": null, "experiment_date": null, "source_update_boundary": "2026-09-06 corrections present; individual table trial dates not stated", "model_weight_precision": "mixed NVFP4/FP8 checkpoint; PLE format varies", "context_limit_tokens": null, "configured_max_requests": null, "input_tokens": null, "output_tokens": null, "actual_output_tokens": null, "concurrency": 1, "cache_state": "page cache dropped before each arm", "thinking": null, "temperature": null, "top_p": null, "draft_tokens": 0, "MTP_main_table_execution": null, "kv_dtype": null, "throughput_repetitions": null, "memory_sampling_instant": null, "reported_size_unit": "GB as labelled by source, byte convention not independently specified", "PLE_format": "INT4 group-16", "workload": "first request, exact request shape unreported; optional resident-vs-mmap cohort, not main 8K/512 table", "memory_backing": "anonymous resident copy" } }, { "label": "INT4 resident sidecar anonymous allocation", "evidence_url": "https://huggingface.co/primitive-ai/Qwen3.8-Flash-Next-PLE-quant/blob/a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763/README.md#L81", "conditions": { "model": "Qwen3.8-Flash-Next", "model_artifact": "primitive-ai/Qwen3.8-Flash-Next-mixed-NVFP4-FP8", "model_revision": null, "PLE_artifact": "primitive-ai/Qwen3.8-Flash-Next-PLE-quant", "PLE_artifact_revision": "a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763", "system": "one RTX PRO 6000 Blackwell 96 GB, 176 GB host RAM, local NVMe", "node_count": 1, "engine": "vLLM with exact-image PLE overlay", "engine_version": null, "runtime_execution_revision": null, "image_tag": "vllm/vllm-openai:qwen38-flash-next", "executed_image_digest": null, "experiment_date": null, "source_update_boundary": "2026-09-06 corrections present; individual table trial dates not stated", "model_weight_precision": "mixed NVFP4/FP8 checkpoint; PLE format varies", "context_limit_tokens": null, "configured_max_requests": 36, "input_tokens": null, "output_tokens": null, "actual_output_tokens": null, "concurrency": null, "cache_state": "prefix-cache-free; mmap page-cache residency not independently logged", "thinking": null, "temperature": null, "top_p": null, "draft_tokens": null, "MTP_main_table_execution": null, "kv_dtype": null, "throughput_repetitions": "two seeds, a/b; not independent boots", "memory_sampling_instant": null, "reported_size_unit": "GB as labelled by source, byte convention not independently specified", "PLE_format": "INT4 group-16", "memory_kind": "anonymous_resident_sidecar_not_total_RSS" } }, { "label": "optional residence cohort KV pool, MTP 0", "evidence_url": "https://huggingface.co/primitive-ai/Qwen3.8-Flash-Next-PLE-quant/blob/a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763/README.md#L80", "conditions": { "model": "Qwen3.8-Flash-Next", "model_artifact": "primitive-ai/Qwen3.8-Flash-Next-mixed-NVFP4-FP8", "model_revision": null, "PLE_artifact": "primitive-ai/Qwen3.8-Flash-Next-PLE-quant", "PLE_artifact_revision": "a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763", "system": "one RTX PRO 6000 Blackwell 96 GB, 176 GB host RAM, local NVMe", "node_count": 1, "engine": "vLLM with exact-image PLE overlay", "engine_version": null, "runtime_execution_revision": null, "image_tag": "vllm/vllm-openai:qwen38-flash-next", "executed_image_digest": null, "experiment_date": null, "source_update_boundary": "2026-09-06 corrections present; individual table trial dates not stated", "model_weight_precision": "mixed NVFP4/FP8 checkpoint; PLE format varies", "context_limit_tokens": null, "configured_max_requests": 36, "input_tokens": null, "output_tokens": null, "actual_output_tokens": null, "concurrency": null, "cache_state": "prefix-cache-free; mmap page-cache residency not independently logged", "thinking": null, "temperature": null, "top_p": null, "draft_tokens": 0, "MTP_main_table_execution": null, "kv_dtype": null, "throughput_repetitions": "two seeds, a/b; not independent boots", "memory_sampling_instant": null, "reported_size_unit": "GB as labelled by source, byte convention not independently specified", "PLE_format": "INT4 group-16", "memory_kind": "allocated_KV_token_pool_not_measured_RAM" } }, { "label": "optional residence cohort KV pool, MTP 3", "evidence_url": "https://huggingface.co/primitive-ai/Qwen3.8-Flash-Next-PLE-quant/blob/a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763/README.md#L80", "conditions": { "model": "Qwen3.8-Flash-Next", "model_artifact": "primitive-ai/Qwen3.8-Flash-Next-mixed-NVFP4-FP8", "model_revision": null, "PLE_artifact": "primitive-ai/Qwen3.8-Flash-Next-PLE-quant", "PLE_artifact_revision": "a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763", "system": "one RTX PRO 6000 Blackwell 96 GB, 176 GB host RAM, local NVMe", "node_count": 1, "engine": "vLLM with exact-image PLE overlay", "engine_version": null, "runtime_execution_revision": null, "image_tag": "vllm/vllm-openai:qwen38-flash-next", "executed_image_digest": null, "experiment_date": null, "source_update_boundary": "2026-09-06 corrections present; individual table trial dates not stated", "model_weight_precision": "mixed NVFP4/FP8 checkpoint; PLE format varies", "context_limit_tokens": null, "configured_max_requests": 36, "input_tokens": null, "output_tokens": null, "actual_output_tokens": null, "concurrency": null, "cache_state": "prefix-cache-free; mmap page-cache residency not independently logged", "thinking": null, "temperature": null, "top_p": null, "draft_tokens": 3, "MTP_main_table_execution": null, "kv_dtype": null, "throughput_repetitions": "two seeds, a/b; not independent boots", "memory_sampling_instant": null, "reported_size_unit": "GB as labelled by source, byte convention not independently specified", "PLE_format": "INT4 group-16", "memory_kind": "allocated_KV_token_pool_not_measured_RAM" } } ], "deployment_options": [], "recipe_parameters": null, "measurement_scope": "source_memory_and_endpoint_observations", "notes": [ "A source document, not an independent experiment count. Preserve measurement type and unknown fields." ], "limitations": [ "No independent Run AI Here model execution.", "Document revision is distinct from execution and weight identity.", "File bytes, host RSS, anonymous memory, MemAvailable, KV allocation and configured context are different quantities." ], "comparison_status": "not_comparable", "editorial_status": "source_derived_detailed_article", "public_benchmark_snapshot": false, "source": { "publisher": "primitive-ai", "url": "https://huggingface.co/primitive-ai/Qwen3.8-Flash-Next-PLE-quant", "repository": "primitive-ai/Qwen3.8-Flash-Next-PLE-quant", "platform": "huggingface", "revision": "a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763", "document_path": "README.md", "pinned_url": "https://huggingface.co/primitive-ai/Qwen3.8-Flash-Next-PLE-quant/blob/a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763/README.md", "report_url": "https://huggingface.co/primitive-ai/Qwen3.8-Flash-Next-PLE-quant/blob/a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763/README.md", "document_sha256": "0085c022b212b241827a09f1a315170737f76298ca5af3bd8a46dea2032faab9", "last_commit_at": "2026-09-14T15:58:31.000Z", "checked_on": "2026-10-18", "metadata_status": "checked" }, "supplemental_sources": [ { "document_id": "qwenple-ples_int4-META.json", "parent_record_id": "qwen38-flash-next-ple-memory", "publisher": "primitive-ai", "repository": "primitive-ai/Qwen3.8-Flash-Next-PLE-quant", "revision": "a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763", "document_path": "ples_int4/META.json", "pinned_url": "https://huggingface.co/primitive-ai/Qwen3.8-Flash-Next-PLE-quant/blob/a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763/ples_int4/META.json", "document_sha256": "e9a1863008450b7ac329d50a3b6b908f21122b8bf5da999bdeff05004161884a", "checked_on": "2026-10-18" }, { "document_id": "qwenple-ples_fp8-META.json", "parent_record_id": "qwen38-flash-next-ple-memory", "publisher": "primitive-ai", "repository": "primitive-ai/Qwen3.8-Flash-Next-PLE-quant", "revision": "a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763", "document_path": "ples_fp8/META.json", "pinned_url": "https://huggingface.co/primitive-ai/Qwen3.8-Flash-Next-PLE-quant/blob/a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763/ples_fp8/META.json", "document_sha256": "de757efa51030ac1e4edb6acdb29507521f8ace860a3be511661a26e5017b632", "checked_on": "2026-10-18" }, { "document_id": "qwenple-ples_nvfp4-META.json", "parent_record_id": "qwen38-flash-next-ple-memory", "publisher": "primitive-ai", "repository": "primitive-ai/Qwen3.8-Flash-Next-PLE-quant", "revision": "a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763", "document_path": "ples_nvfp4/META.json", "pinned_url": "https://huggingface.co/primitive-ai/Qwen3.8-Flash-Next-PLE-quant/blob/a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763/ples_nvfp4/META.json", "document_sha256": "5aa4d53703f67c8c664d674c95f16c29c8ce7a94a2d7d9ac27ae8afe4c0b56ef", "checked_on": "2026-10-18" } ], "license": { "declared_spdx": "apache-2.0", "scope": "Source declaration only; base weights, converters, runtimes and source integration material retain their own terms." }, "editorial": { "version": 1, "published_on": "2026-10-18", "modified_on": "2026-10-18", "byline": "Run AI Here editorial", "creation_method": "Original source-grounded editorial synthesis with local Spark Qwen assistance and Codex review; no site model benchmark.", "answer": "This experiment changes the n-gram PLE table backing, rather than quantizing every model component. The source compares a BF16 table held in RAM with file-mapped sidecars on one 96 GB RTX PRO 6000 and a 176 GB host. Main-table throughput uses 8K input, 512 requested output tokens and no prefix-cache hits. Full checkpoint revision and executed image digest are not supplied.", "sections": [ { "heading": "Which allocation is reduced?", "paragraphs": [ "This experiment changes the n-gram PLE table backing, rather than quantizing every model component. The source compares a BF16 table held in RAM with file-mapped sidecars on one 96 GB RTX PRO 6000 and a 176 GB host. Main-table throughput uses 8K input, 512 requested output tokens and no prefix-cache hits. Full checkpoint revision and executed image digest are not supplied." ], "citations": [ "https://huggingface.co/primitive-ai/Qwen3.8-Flash-Next-PLE-quant/blob/a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763/README.md", "https://huggingface.co/primitive-ai/Qwen3.8-Flash-Next-PLE-quant/blob/a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763/ples_int4/META.json" ] }, { "heading": "Why is RSS different from anonymous memory?", "paragraphs": [ "The selected BF16 and INT4 rows retain artifact size separately from host RSS. File-mapped RSS includes reclaimable page-cache pages; it is not a permanent anonymous allocation. The optional INT4 resident copy instead adds a reported 29.8 GiB of anonymous sidecar memory, above the worker allocation. Source-labelled GB and GiB are not silently equated." ], "citations": [ "https://huggingface.co/primitive-ai/Qwen3.8-Flash-Next-PLE-quant/blob/a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763/README.md" ] }, { "heading": "What remains outside the evidence?", "paragraphs": [ "The optional cold-residence cohort drops page cache before each arm and has a different, incompletely reported request shape. Its first-request speed is not a warm throughput average. MTP changes the allocated KV token pool; mmap versus resident does not change that pool within each reported MTP setting. Neither the table-size reduction nor suite-level sanity proves general quality equivalence." ], "citations": [ "https://huggingface.co/primitive-ai/Qwen3.8-Flash-Next-PLE-quant/blob/a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763/README.md" ] }, { "heading": "Does the native window fit every launch configuration?", "paragraphs": [ "The source reports a native-window startup failure with MTP when 7.57 GiB of KV was required but 5.95 GiB remained. Its selected main table used a shorter 32768-token launch window, while a separate field report used 135168. These are configuration-specific observations, not a universal maximum or evidence that every format fits the native window." ], "citations": [ "https://huggingface.co/primitive-ai/Qwen3.8-Flash-Next-PLE-quant/blob/a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763/README.md" ] } ], "facts": [], "reported_results": [ { "label": "BF16 in RAM table size", "metric": "artifact_size", "value": 95.4, "unit": "GB", "statistic": "source-reported PLE table size, not whole model or peak RAM", "uncertainty": null, "conditions": { "model": "Qwen3.8-Flash-Next", "model_artifact": "primitive-ai/Qwen3.8-Flash-Next-mixed-NVFP4-FP8", "model_revision": null, "PLE_artifact": "primitive-ai/Qwen3.8-Flash-Next-PLE-quant", "PLE_artifact_revision": "a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763", "system": "one RTX PRO 6000 Blackwell 96 GB, 176 GB host RAM, local NVMe", "node_count": 1, "engine": "vLLM with exact-image PLE overlay", "engine_version": null, "runtime_execution_revision": null, "image_tag": "vllm/vllm-openai:qwen38-flash-next", "executed_image_digest": null, "experiment_date": null, "source_update_boundary": "2026-09-06 corrections present; individual table trial dates not stated", "model_weight_precision": "mixed NVFP4/FP8 checkpoint; PLE format varies", "context_limit_tokens": 32768, "configured_max_requests": 36, "input_tokens": 8192, "output_tokens": 512, "actual_output_tokens": null, "concurrency": 1, "cache_state": "prefix-cache-free; mmap page-cache residency not independently logged", "thinking": true, "temperature": null, "top_p": null, "draft_tokens": null, "MTP_main_table_execution": null, "kv_dtype": null, "throughput_repetitions": "two seeds, a/b; not independent boots", "memory_sampling_instant": null, "reported_size_unit": "GB as labelled by source, byte convention not independently specified", "PLE_format": "BF16 in RAM", "memory_kind": "PLE_artifact_size", "memory_backing": "in RAM" }, "evidence_state": "source_reported", "site_reproduced": false, "source_document_id": "qwenple-readme", "source_document_sha256": "0085c022b212b241827a09f1a315170737f76298ca5af3bd8a46dea2032faab9", "source_line": 96, "evidence_url": "https://huggingface.co/primitive-ai/Qwen3.8-Flash-Next-PLE-quant/blob/a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763/README.md#L96" }, { "label": "BF16 in RAM host RSS", "metric": "host_rss", "value": 95, "unit": "GB", "statistic": "approximate source host RSS", "uncertainty": { "approximate": true }, "conditions": { "model": "Qwen3.8-Flash-Next", "model_artifact": "primitive-ai/Qwen3.8-Flash-Next-mixed-NVFP4-FP8", "model_revision": null, "PLE_artifact": "primitive-ai/Qwen3.8-Flash-Next-PLE-quant", "PLE_artifact_revision": "a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763", "system": "one RTX PRO 6000 Blackwell 96 GB, 176 GB host RAM, local NVMe", "node_count": 1, "engine": "vLLM with exact-image PLE overlay", "engine_version": null, "runtime_execution_revision": null, "image_tag": "vllm/vllm-openai:qwen38-flash-next", "executed_image_digest": null, "experiment_date": null, "source_update_boundary": "2026-09-06 corrections present; individual table trial dates not stated", "model_weight_precision": "mixed NVFP4/FP8 checkpoint; PLE format varies", "context_limit_tokens": 32768, "configured_max_requests": 36, "input_tokens": 8192, "output_tokens": 512, "actual_output_tokens": null, "concurrency": 1, "cache_state": "prefix-cache-free; mmap page-cache residency not independently logged", "thinking": true, "temperature": null, "top_p": null, "draft_tokens": null, "MTP_main_table_execution": null, "kv_dtype": null, "throughput_repetitions": "two seeds, a/b; not independent boots", "memory_sampling_instant": null, "reported_size_unit": "GB as labelled by source, byte convention not independently specified", "PLE_format": "BF16 in RAM", "memory_kind": "host_RSS_not_necessarily_anonymous", "memory_backing": "in RAM" }, "evidence_state": "source_reported", "site_reproduced": false, "source_document_id": "qwenple-readme", "source_document_sha256": "0085c022b212b241827a09f1a315170737f76298ca5af3bd8a46dea2032faab9", "source_line": 96, "evidence_url": "https://huggingface.co/primitive-ai/Qwen3.8-Flash-Next-PLE-quant/blob/a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763/README.md#L96" }, { "label": "INT4 group-16 mmap table size", "metric": "artifact_size", "value": 32, "unit": "GB", "statistic": "source-reported PLE table size, not whole model or peak RAM", "uncertainty": null, "conditions": { "model": "Qwen3.8-Flash-Next", "model_artifact": "primitive-ai/Qwen3.8-Flash-Next-mixed-NVFP4-FP8", "model_revision": null, "PLE_artifact": "primitive-ai/Qwen3.8-Flash-Next-PLE-quant", "PLE_artifact_revision": "a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763", "system": "one RTX PRO 6000 Blackwell 96 GB, 176 GB host RAM, local NVMe", "node_count": 1, "engine": "vLLM with exact-image PLE overlay", "engine_version": null, "runtime_execution_revision": null, "image_tag": "vllm/vllm-openai:qwen38-flash-next", "executed_image_digest": null, "experiment_date": null, "source_update_boundary": "2026-09-06 corrections present; individual table trial dates not stated", "model_weight_precision": "mixed NVFP4/FP8 checkpoint; PLE format varies", "context_limit_tokens": 32768, "configured_max_requests": 36, "input_tokens": 8192, "output_tokens": 512, "actual_output_tokens": null, "concurrency": 1, "cache_state": "prefix-cache-free; mmap page-cache residency not independently logged", "thinking": true, "temperature": null, "top_p": null, "draft_tokens": null, "MTP_main_table_execution": null, "kv_dtype": null, "throughput_repetitions": "two seeds, a/b; not independent boots", "memory_sampling_instant": null, "reported_size_unit": "GB as labelled by source, byte convention not independently specified", "PLE_format": "INT4 group-16 mmap", "memory_kind": "PLE_artifact_size", "memory_backing": "reclaimable mapped file pages" }, "evidence_state": "source_reported", "site_reproduced": false, "source_document_id": "qwenple-readme", "source_document_sha256": "0085c022b212b241827a09f1a315170737f76298ca5af3bd8a46dea2032faab9", "source_line": 98, "evidence_url": "https://huggingface.co/primitive-ai/Qwen3.8-Flash-Next-PLE-quant/blob/a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763/README.md#L98" }, { "label": "INT4 group-16 mmap host RSS", "metric": "host_rss", "value": 32.9, "unit": "GB", "statistic": "source host RSS including reclaimable file pages", "uncertainty": null, "conditions": { "model": "Qwen3.8-Flash-Next", "model_artifact": "primitive-ai/Qwen3.8-Flash-Next-mixed-NVFP4-FP8", "model_revision": null, "PLE_artifact": "primitive-ai/Qwen3.8-Flash-Next-PLE-quant", "PLE_artifact_revision": "a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763", "system": "one RTX PRO 6000 Blackwell 96 GB, 176 GB host RAM, local NVMe", "node_count": 1, "engine": "vLLM with exact-image PLE overlay", "engine_version": null, "runtime_execution_revision": null, "image_tag": "vllm/vllm-openai:qwen38-flash-next", "executed_image_digest": null, "experiment_date": null, "source_update_boundary": "2026-09-06 corrections present; individual table trial dates not stated", "model_weight_precision": "mixed NVFP4/FP8 checkpoint; PLE format varies", "context_limit_tokens": 32768, "configured_max_requests": 36, "input_tokens": 8192, "output_tokens": 512, "actual_output_tokens": null, "concurrency": 1, "cache_state": "prefix-cache-free; mmap page-cache residency not independently logged", "thinking": true, "temperature": null, "top_p": null, "draft_tokens": null, "MTP_main_table_execution": null, "kv_dtype": null, "throughput_repetitions": "two seeds, a/b; not independent boots", "memory_sampling_instant": null, "reported_size_unit": "GB as labelled by source, byte convention not independently specified", "PLE_format": "INT4 group-16 mmap", "memory_kind": "host_RSS_not_necessarily_anonymous", "memory_backing": "reclaimable mapped file pages" }, "evidence_state": "source_reported", "site_reproduced": false, "source_document_id": "qwenple-readme", "source_document_sha256": "0085c022b212b241827a09f1a315170737f76298ca5af3bd8a46dea2032faab9", "source_line": 98, "evidence_url": "https://huggingface.co/primitive-ai/Qwen3.8-Flash-Next-PLE-quant/blob/a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763/README.md#L98" }, { "label": "mmap cold first request", "metric": "decode_rate", "value": 19.0, "unit": "tokens/s", "statistic": "first-request observation; warm asymptote is a separate quantity", "uncertainty": null, "conditions": { "model": "Qwen3.8-Flash-Next", "model_artifact": "primitive-ai/Qwen3.8-Flash-Next-mixed-NVFP4-FP8", "model_revision": null, "PLE_artifact": "primitive-ai/Qwen3.8-Flash-Next-PLE-quant", "PLE_artifact_revision": "a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763", "system": "one RTX PRO 6000 Blackwell 96 GB, 176 GB host RAM, local NVMe", "node_count": 1, "engine": "vLLM with exact-image PLE overlay", "engine_version": null, "runtime_execution_revision": null, "image_tag": "vllm/vllm-openai:qwen38-flash-next", "executed_image_digest": null, "experiment_date": null, "source_update_boundary": "2026-09-06 corrections present; individual table trial dates not stated", "model_weight_precision": "mixed NVFP4/FP8 checkpoint; PLE format varies", "context_limit_tokens": null, "configured_max_requests": null, "input_tokens": null, "output_tokens": null, "actual_output_tokens": null, "concurrency": 1, "cache_state": "page cache dropped before each arm", "thinking": null, "temperature": null, "top_p": null, "draft_tokens": 0, "MTP_main_table_execution": null, "kv_dtype": null, "throughput_repetitions": null, "memory_sampling_instant": null, "reported_size_unit": "GB as labelled by source, byte convention not independently specified", "PLE_format": "INT4 group-16", "workload": "first request, exact request shape unreported; optional resident-vs-mmap cohort, not main 8K/512 table", "memory_backing": "NVMe mmap" }, "evidence_state": "source_reported", "site_reproduced": false, "source_document_id": "qwenple-readme", "source_document_sha256": "0085c022b212b241827a09f1a315170737f76298ca5af3bd8a46dea2032faab9", "source_line": 68, "evidence_url": "https://huggingface.co/primitive-ai/Qwen3.8-Flash-Next-PLE-quant/blob/a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763/README.md#L68" }, { "label": "resident cold first request", "metric": "decode_rate", "value": 74.3, "unit": "tokens/s", "statistic": "first-request observation; warm asymptote is a separate quantity", "uncertainty": null, "conditions": { "model": "Qwen3.8-Flash-Next", "model_artifact": "primitive-ai/Qwen3.8-Flash-Next-mixed-NVFP4-FP8", "model_revision": null, "PLE_artifact": "primitive-ai/Qwen3.8-Flash-Next-PLE-quant", "PLE_artifact_revision": "a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763", "system": "one RTX PRO 6000 Blackwell 96 GB, 176 GB host RAM, local NVMe", "node_count": 1, "engine": "vLLM with exact-image PLE overlay", "engine_version": null, "runtime_execution_revision": null, "image_tag": "vllm/vllm-openai:qwen38-flash-next", "executed_image_digest": null, "experiment_date": null, "source_update_boundary": "2026-09-06 corrections present; individual table trial dates not stated", "model_weight_precision": "mixed NVFP4/FP8 checkpoint; PLE format varies", "context_limit_tokens": null, "configured_max_requests": null, "input_tokens": null, "output_tokens": null, "actual_output_tokens": null, "concurrency": 1, "cache_state": "page cache dropped before each arm", "thinking": null, "temperature": null, "top_p": null, "draft_tokens": 0, "MTP_main_table_execution": null, "kv_dtype": null, "throughput_repetitions": null, "memory_sampling_instant": null, "reported_size_unit": "GB as labelled by source, byte convention not independently specified", "PLE_format": "INT4 group-16", "workload": "first request, exact request shape unreported; optional resident-vs-mmap cohort, not main 8K/512 table", "memory_backing": "anonymous resident copy" }, "evidence_state": "source_reported", "site_reproduced": false, "source_document_id": "qwenple-readme", "source_document_sha256": "0085c022b212b241827a09f1a315170737f76298ca5af3bd8a46dea2032faab9", "source_line": 69, "evidence_url": "https://huggingface.co/primitive-ai/Qwen3.8-Flash-Next-PLE-quant/blob/a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763/README.md#L69" }, { "label": "INT4 resident sidecar anonymous allocation", "metric": "anonymous_memory", "value": 29.8, "unit": "GiB", "statistic": "sidecar allocation only, on top of roughly 8 GiB held by worker", "uncertainty": null, "conditions": { "model": "Qwen3.8-Flash-Next", "model_artifact": "primitive-ai/Qwen3.8-Flash-Next-mixed-NVFP4-FP8", "model_revision": null, "PLE_artifact": "primitive-ai/Qwen3.8-Flash-Next-PLE-quant", "PLE_artifact_revision": "a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763", "system": "one RTX PRO 6000 Blackwell 96 GB, 176 GB host RAM, local NVMe", "node_count": 1, "engine": "vLLM with exact-image PLE overlay", "engine_version": null, "runtime_execution_revision": null, "image_tag": "vllm/vllm-openai:qwen38-flash-next", "executed_image_digest": null, "experiment_date": null, "source_update_boundary": "2026-09-06 corrections present; individual table trial dates not stated", "model_weight_precision": "mixed NVFP4/FP8 checkpoint; PLE format varies", "context_limit_tokens": null, "configured_max_requests": 36, "input_tokens": null, "output_tokens": null, "actual_output_tokens": null, "concurrency": null, "cache_state": "prefix-cache-free; mmap page-cache residency not independently logged", "thinking": null, "temperature": null, "top_p": null, "draft_tokens": null, "MTP_main_table_execution": null, "kv_dtype": null, "throughput_repetitions": "two seeds, a/b; not independent boots", "memory_sampling_instant": null, "reported_size_unit": "GB as labelled by source, byte convention not independently specified", "PLE_format": "INT4 group-16", "memory_kind": "anonymous_resident_sidecar_not_total_RSS" }, "evidence_state": "source_reported", "site_reproduced": false, "source_document_id": "qwenple-readme", "source_document_sha256": "0085c022b212b241827a09f1a315170737f76298ca5af3bd8a46dea2032faab9", "source_line": 81, "evidence_url": "https://huggingface.co/primitive-ai/Qwen3.8-Flash-Next-PLE-quant/blob/a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763/README.md#L81" }, { "label": "optional residence cohort KV pool, MTP 0", "metric": "kv_pool_tokens", "value": 426639, "unit": "tokens", "statistic": "reported identical mmap/resident allocation for this MTP setting", "uncertainty": null, "conditions": { "model": "Qwen3.8-Flash-Next", "model_artifact": "primitive-ai/Qwen3.8-Flash-Next-mixed-NVFP4-FP8", "model_revision": null, "PLE_artifact": "primitive-ai/Qwen3.8-Flash-Next-PLE-quant", "PLE_artifact_revision": "a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763", "system": "one RTX PRO 6000 Blackwell 96 GB, 176 GB host RAM, local NVMe", "node_count": 1, "engine": "vLLM with exact-image PLE overlay", "engine_version": null, "runtime_execution_revision": null, "image_tag": "vllm/vllm-openai:qwen38-flash-next", "executed_image_digest": null, "experiment_date": null, "source_update_boundary": "2026-09-06 corrections present; individual table trial dates not stated", "model_weight_precision": "mixed NVFP4/FP8 checkpoint; PLE format varies", "context_limit_tokens": null, "configured_max_requests": 36, "input_tokens": null, "output_tokens": null, "actual_output_tokens": null, "concurrency": null, "cache_state": "prefix-cache-free; mmap page-cache residency not independently logged", "thinking": null, "temperature": null, "top_p": null, "draft_tokens": 0, "MTP_main_table_execution": null, "kv_dtype": null, "throughput_repetitions": "two seeds, a/b; not independent boots", "memory_sampling_instant": null, "reported_size_unit": "GB as labelled by source, byte convention not independently specified", "PLE_format": "INT4 group-16", "memory_kind": "allocated_KV_token_pool_not_measured_RAM" }, "evidence_state": "source_reported", "site_reproduced": false, "source_document_id": "qwenple-readme", "source_document_sha256": "0085c022b212b241827a09f1a315170737f76298ca5af3bd8a46dea2032faab9", "source_line": 80, "evidence_url": "https://huggingface.co/primitive-ai/Qwen3.8-Flash-Next-PLE-quant/blob/a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763/README.md#L80" }, { "label": "optional residence cohort KV pool, MTP 3", "metric": "kv_pool_tokens", "value": 134192, "unit": "tokens", "statistic": "reported identical mmap/resident allocation for this MTP setting", "uncertainty": null, "conditions": { "model": "Qwen3.8-Flash-Next", "model_artifact": "primitive-ai/Qwen3.8-Flash-Next-mixed-NVFP4-FP8", "model_revision": null, "PLE_artifact": "primitive-ai/Qwen3.8-Flash-Next-PLE-quant", "PLE_artifact_revision": "a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763", "system": "one RTX PRO 6000 Blackwell 96 GB, 176 GB host RAM, local NVMe", "node_count": 1, "engine": "vLLM with exact-image PLE overlay", "engine_version": null, "runtime_execution_revision": null, "image_tag": "vllm/vllm-openai:qwen38-flash-next", "executed_image_digest": null, "experiment_date": null, "source_update_boundary": "2026-09-06 corrections present; individual table trial dates not stated", "model_weight_precision": "mixed NVFP4/FP8 checkpoint; PLE format varies", "context_limit_tokens": null, "configured_max_requests": 36, "input_tokens": null, "output_tokens": null, "actual_output_tokens": null, "concurrency": null, "cache_state": "prefix-cache-free; mmap page-cache residency not independently logged", "thinking": null, "temperature": null, "top_p": null, "draft_tokens": 3, "MTP_main_table_execution": null, "kv_dtype": null, "throughput_repetitions": "two seeds, a/b; not independent boots", "memory_sampling_instant": null, "reported_size_unit": "GB as labelled by source, byte convention not independently specified", "PLE_format": "INT4 group-16", "memory_kind": "allocated_KV_token_pool_not_measured_RAM" }, "evidence_state": "source_reported", "site_reproduced": false, "source_document_id": "qwenple-readme", "source_document_sha256": "0085c022b212b241827a09f1a315170737f76298ca5af3bd8a46dea2032faab9", "source_line": 80, "evidence_url": "https://huggingface.co/primitive-ai/Qwen3.8-Flash-Next-PLE-quant/blob/a0fa93f2b9ed5fcab0cf01b60e88ff1ce13a4763/README.md#L80" } ], "source_document_sha256": "0085c022b212b241827a09f1a315170737f76298ca5af3bd8a46dea2032faab9" }, "memory_observations": [ { "record_id": "qwen38-flash-next-ple-memory", "kind": "source_main_table_host_RSS", "BF16_in_RAM_GB_approx": 95, "INT4_mmap_GB": 32.9, "mapped_pages_reclaimable": true, "anonymous_memory_inferred_from_RSS": false, "GB_byte_convention_verified": false }, { "record_id": "qwen38-flash-next-ple-memory", "kind": "optional_anonymous_residence", "INT4_sidecar_GiB": 29.8, "additional_worker_GiB_approx": 8, "same_cohort_as_main_table": false } ] }