GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
Jiunsong
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
Jiunsong
Model Tree
deepseek-ai/DeepSeek-V4-Flash-0731
Input Modalities
Output Modalities
Supported Functionality
| Architecture | DeepSeek V4 Flash, 304B-class MoE, 43 backbone + 3 MTP layers, 256 routed experts, top-6 |
| Checkpoint | Complete Transformers/safetensors checkpoint, about 169.5 GB, 48 parent shards plus fused SuperDeepseek overlays |
| Native precision | FP4 experts + FP8 E4M3 block-quantized paths + BF16 quality-sensitive paths |
| Targeted update | 46 attn.wo_b weight/scale pairs plus one bounded BF16 output-head recovery |
| Context | 1,048,576 configured; 1,028,621 actual prompt tokens accepted and retrieved |
| Decode | 123.3459 forced / 123.2819 structured-tool aggregate tok/s at p256/C6 |
| Behavior | Worst-mode refusal 97.92% -> 4.17%; tool gates 100% |
This is the native-format full checkpoint, not a full-BF16 reconstruction. DeepSeek's official V4 Flash checkpoint is itself hybrid precision; no authentic all-BF16 upstream checkpoint is published. This release deliberately preserves that official representation instead of dequantizing it and presenting approximated BF16 weights as “original.”
| Component | Precision / storage |
|---|---|
| MoE expert weights | FP4, inherited from the official expert_dtype=fp4 layout |
| Block-quantized paths | FP8 E4M3, dynamic activation scaling, 128x128 blocks, UE8M0 scales |
43 backbone + 3 MTP attn.wo_b updates | Deterministic FP8 weight/scale overlay, 92 tensors |
| Default unquantized and quality-sensitive paths | BF16 (torch_dtype=bfloat16), with upstream F32 metadata/normalization where defined |
| Output-head recovery | One bounded BF16 head.weight overlay, rank 64, relative Frobenius delta 0.0025 |
The parent is pinned to deepseek-ai/DeepSeek-V4-Flash-0731@9e165c30e2704aec5d9d593cce3eebd58bbef1cb. All 48 parent shard identities were checked against that revision.
For the release name and scripts tuned specifically for a two-node DGX Spark deployment, see Jiunsong/SuperDeepseek-V4-Flash-abliterated-MQ-2xDGX. Both repositories contain the same released SuperDeepseek weights; this repository is the general/native-layout edition, while the sibling makes the measured MQ + 2xDGX deployment target explicit in its name.
The release uses two measured refusal-subspace passes:
The final checkpoint changes only the 43 backbone and three MTP attn.wo_b weight/scale pairs plus the bounded output head. The routed experts remain untouched.
| Metric | Official parent | SuperDeepseek |
|---|---|---|
| Worst-mode refusal | 97.92% | 4.17% |
| Worst empty answer | 0.00% | 0.00% |
| Worst tool compliance | 100.00% | 100.00% |
| Worst correct-tool rate | 100.00% | 100.00% |
| Minimum capability mean | 0.9375 | 0.9583 |
The independently reloaded checkpoint reproduced the selected candidate's deterministic validation behavior. The release gate covered empty output, Unicode, repetition, serialization, reasoning, code, formatting, tool selection, and structured tool calls.
These are aggregate concurrent post-first-token decode rates, not single-stream numbers. Six distinct prompts, fixed-length generation, and the sealed sparkDash-style p256/C6 contract were used.
| Workload | Prompt / concurrency | Three trials | Median |
|---|---|---|---|
| Forced output | p256 / C6 | 117.7944, 123.3459, 125.9469 tok/s | 123.3459 tok/s |
| Structured tool output | p256 / C6 | 123.6285, 123.2819, 117.1843 tok/s | 123.2819 tok/s |
The benchmark follows sparkDash commit dfde4214f32b174880832a4d317d3c0567750ac5. Hardware and runtime were 2x NVIDIA DGX Spark / GB10, TP=2, direct CX-7 RoCEv2, vLLM, FlashInfer b12x MoE, DSpark K=1 speculative decoding, and NVFP4 DS-MLA KV cache. Hardware-specific figures should not be assumed on other systems.
| Actual prompt tokens | Accepted | Needle retrieved |
|---|---|---|
| 149,845 | Yes | Yes |
| 1,028,621 | Yes | Yes |
The configured maximum is 1,048,576 tokens. These are end-to-end acceptance and retrieval probes, not a claim of perfect recall for every task or needle position.
The verified profile uses TP=2 over direct CX-7 RoCEv2 with the ghcr.io/anemll/dspark-vllm-gx10:0.1.1 runtime:
bash
vllm serve /model \--served-model-name SuperDeepseek-V4-Flash-abliterated \--tensor-parallel-size 2 \--max-model-len 1048576 \--kv-cache-dtype nvfp4_ds_mla \--moe-backend flashinfer_b12x \--enable-prefix-caching \--async-scheduling \--enable-chunked-prefill \--speculative-config '{"method":"dspark","num_speculative_tokens":1,"draft_sample_method":"greedy"}'
The exact two-rank launcher and machine-readable release evidence are included under repro/ and evidence/.
9e165c30e2704aec5d9d593cce3eebd58bbef1cbattn.wo_b weight/scale pairs| Artifact | SHA-256 |
|---|---|
| SuperDeepseek FP8 overlay | 86c8494d0b02a01ccb4d4de5ad48a66d26f56341e1dd3eb17eb61b40710f73ba |
| BF16 head-recovery overlay | 3d49e3e05ba864666054328f475058a9558603aac70d1a191b352be65bce1428 |
| Sealed throughput report | 1cd7734205f4d03cddf241d5ed2a5f423f402d194222c9abc3bdacc965cce370 |
| 1M-context evidence | 6745453ee65b9e59581d715bf668bc710b0bc680fe726763e9a750dd25f45565 |
“Abliterated” describes a measured reduction of the selected refusal subspace. It does not make every answer correct, remove the need for deployment controls, or transfer responsibility away from the operator. The capability suites are finite regression gates, and the upstream model's license and limitations continue to apply.
MIT, following the upstream DeepSeek V4 Flash release.
| Verified serving KV cache | NVFP4 DS-MLA |