Hanrui / sglang /docs /advanced_features /speculative_decoding.md

Add files using upload-large-folder tool

a227c91 verified 24 days ago

25.6 kB

Speculative Decoding

SGLang provides several speculative decoding options, including EAGLE-2/EAGLE-3, MTP, classic draft-model decoding, and an NGRAM-based variant. Our implementation aims to maximize speed and efficiency and is considered to be among the fastest in open-source LLM engines.

Summary

Jump to sections

EAGLE Decoding
Multi Token Prediction
Standalone Speculative Decoding (Small Draft Model)
Speculative Decoding V2 (Overlap Scheduler)
Ngram Speculative Decoding
Full Parameter Reference
OOM Troubleshooting
References

Quick guidance

Best speed/quality (recommended): Use EAGLE-3 with --speculative-algorithm EAGLE3.
Strong default / broad compatibility: Use EAGLE-2 with --speculative-algorithm EAGLE.
Lower lm_head overhead for EAGLE-2: Enable FR-Spec with --speculative-token-map.
Model is MTP-enabled: Use MTP via speculative decoding (often with small speculative_num_steps/topk/num_draft_tokens, see the example section).
You have a smaller draft LLM: Use STANDALONE (--speculative-algorithm STANDALONE).
No extra model available: Use NGRAM (--speculative-algorithm NGRAM, CUDA-only).
Want overlap scheduler (experimental): Enable SpecV2 with SGLANG_ENABLE_SPEC_V2=True (requires --speculative-eagle-topk 1).

Method comparison (mini table)

Method	Draft source	Separate draft model?	How to enable	Notes / constraints
EAGLE-2	EAGLE draft model (feature drafting + tree)	Typically yes	`--speculative-algorithm EAGLE` + `--speculative-draft-model-path ...`	Tune `--speculative-num-steps`, `--speculative-eagle-topk`, `--speculative-num-draft-tokens`
EAGLE-2 + `torch.compile`	Same as EAGLE-2	Typically yes	Add `--enable-torch-compile` (optionally `--torch-compile-max-bs`)	Benefit varies by hardware/model; benchmark to verify
EAGLE-2 + FR-Spec	Same as EAGLE-2 + token subset	Typically yes	Add `--speculative-token-map ...`	Reduces `lm_head` overhead with high-frequency token vocab
EAGLE-3	EAGLE3 draft model	Yes	`--speculative-algorithm EAGLE3` + `--speculative-draft-model-path ...`	Best throughput in the benchmark below
MTP	Built-in multi-token heads (model-specific)	Often no	See Multi Token Prediction section	Uses speculative workflow; draft path may be auto-handled for some models
STANDALONE	Smaller draft LLM (token-level)	Yes	`--speculative-algorithm STANDALONE` + `--speculative-draft-model-path ...`	Does not support `--enable-dp-attention`
SpecV2 (experimental)	V2 workers + overlap scheduler	N/A	`SGLANG_ENABLE_SPEC_V2=True`	Only supports `--speculative-eagle-topk 1`; applies to `EAGLE`, `EAGLE3`, `STANDALONE`
NGRAM	Ngram cache from previous tokens	No	`--speculative-algorithm NGRAM`	CUDA-only; no `--enable-dp-attention`; disables overlap scheduler & mixed chunked prefill

Performance Highlights

Please see below for the huge improvements on throughput for LLaMA-Instruct 3.1 8B tested on MT bench that can be achieved via EAGLE3 decoding. For further details please see the EAGLE3 paper.

Method	Throughput (tokens/s)
SGLang (w/o speculative, 1x H100)	158.34 tokens/s
SGLang + EAGLE-2 (1x H100)	244.10 tokens/s
SGLang + EAGLE-3 (1x H100)	373.25 tokens/s

EAGLE Decoding

To enable EAGLE speculative decoding the following parameters are relevant:

Parameter	Description	Default
`--speculative-draft-model-path`	Draft model path/weights. Typically required for EAGLE/EAGLE3 and STANDALONE. For some MTP-enabled models, this can be omitted.	`None`
`--speculative-num-steps`	Depth of autoregressive drafting. Increases speculation range but risks rejection cascades.	Auto (`5` for Llama/Grok; `3` for many other models)
`--speculative-eagle-topk`	Branching factor per step. Improves candidate diversity and acceptance rate, but increases memory/compute consumption.	Auto (`4` for Llama/Grok; `1` for many other models)
`--speculative-num-draft-tokens`	Maximum parallel verification capacity. Allows deeper tree evaluation but increases GPU memory usage.	Auto (`8` for Llama/Grok; `4` for many other models). If `topk=1`, it is adjusted to `num_steps + 1`.
`--speculative-accept-threshold-single`	Acceptance threshold for single-token verification. Lower values accept more aggressively.	`1.0`
`--speculative-accept-threshold-acc`	Accumulated acceptance threshold across steps.	`1.0`
`--speculative-attention-mode`	Attention mode for speculative operations (`prefill` or `decode`), affecting both target verification and draft extension.	`"prefill"`
`--speculative-draft-attention-backend`	Override attention backend for the draft model.	`None` (same as target)
`--speculative-draft-model-quantization`	Quantization method for the draft model. Use `"unquant"` to force no quantization even when the target model is quantized.	Same as target model
`--speculative-draft-model-revision`	Specific revision/commit of the draft model to load.	`None` (auto-set to `"main"` when `--speculative-draft-model-path` is set and revision is omitted)
`--speculative-draft-load-format`	Load format for the draft model weights.	`None`

These parameters are mostly the same for EAGLE-2 and EAGLE-3. --speculative-token-map is ignored for EAGLE-3 models. For --speculative-num-steps, --speculative-eagle-topk, and --speculative-num-draft-tokens: leave all three unset to use auto-tuning, or set all three explicitly when tuning.

You can find the best combinations of these parameters with bench_speculative.py.

EAGLE-2 Decoding

You can enable EAGLE-2 Decoding by setting --speculative-algorithm EAGLE and choosing an appropriate model.

Launch the server:

python3 -m sglang.launch_server \
    --model meta-llama/Llama-2-7b-chat-hf \
    --speculative-algorithm EAGLE \
    --speculative-draft-model-path lmsys/sglang-EAGLE-llama2-chat-7B \
    --speculative-num-steps 3 \
    --speculative-eagle-topk 4 \
    --speculative-num-draft-tokens 16 \
    --mem-fraction-static 0.7 \
    --cuda-graph-max-bs 8 \
    --log-level warning

Send a request:

import openai

client = openai.Client(base_url="http://127.0.0.1:30000/v1", api_key="None")

response = client.chat.completions.create(
    model="meta-llama/Llama-2-7b-chat-hf",
    messages=[
        {"role": "user", "content": "List 3 countries and their capitals."},
    ],
    temperature=0,
    max_tokens=64,
)

print(response.choices[0].message.content)

EAGLE-2 Decoding with `torch.compile`

You can optionally enable torch.compile to apply kernel-level optimizations (operator fusion, autotune) to the draft model. The actual speedup depends on your hardware, model architecture, and batch size. In some configurations (e.g., small draft models on H100 where cuBLAS is already optimal and CUDA graphs are enabled), the benefit may be negligible. We recommend benchmarking with and without this flag on your specific setup to verify whether it helps.

To enable it, add --enable-torch-compile and optionally set --torch-compile-max-bs:

python3 -m sglang.launch_server \
    --model meta-llama/Llama-2-7b-chat-hf \
    --speculative-algorithm EAGLE \
    --speculative-draft-model-path lmsys/sglang-EAGLE-llama2-chat-7B \
    --speculative-num-steps 3 \
    --speculative-eagle-topk 4 \
    --speculative-num-draft-tokens 16 \
    --mem-fraction-static 0.7 \
    --enable-torch-compile \
    --torch-compile-max-bs 8 \
    --log-level warning

Send a request:

import openai

client = openai.Client(base_url="http://127.0.0.1:30000/v1", api_key="None")

response = client.chat.completions.create(
    model="meta-llama/Llama-2-7b-chat-hf",
    messages=[
        {"role": "user", "content": "List 3 countries and their capitals."},
    ],
    temperature=0,
    max_tokens=64,
)

print(response.choices[0].message.content)

EAGLE-2 Decoding via Frequency-Ranked Speculative Sampling

By employing a truncated high-frequency token vocabulary in the draft model, EAGLE speculative decoding reduces lm_head computational overhead while accelerating the pipeline without quality degradation. For more details, check out the paper.

In our implementation, set --speculative-token-map to enable the optimization. You can get the high-frequency tokens in FR-Spec from this model. Or you can obtain high-frequency tokens by directly downloading these tokens from this repo.

Thanks for the contribution from Weilin Zhao and Zhousx.

python3 -m sglang.launch_server \
    --model meta-llama/Meta-Llama-3-8B-Instruct \
    --speculative-algorithm EAGLE \
    --speculative-draft-model-path lmsys/sglang-EAGLE-LLaMA3-Instruct-8B \
    --speculative-num-steps 3 \
    --speculative-eagle-topk 4 \
    --speculative-num-draft-tokens 16 \
    --speculative-token-map thunlp/LLaMA3-Instruct-8B-FR-Spec/freq_32768.pt \
    --mem-fraction-static 0.7 \
    --cuda-graph-max-bs 8 \
    --dtype float16 \
    --log-level warning

Send a request:

import openai

client = openai.Client(base_url="http://127.0.0.1:30000/v1", api_key="None")

response = client.chat.completions.create(
    model="meta-llama/Meta-Llama-3-8B-Instruct",
    messages=[
        {"role": "user", "content": "List 3 countries and their capitals."},
    ],
    temperature=0,
    max_tokens=64,
)

print(response.choices[0].message.content)

EAGLE-3 Decoding

You can enable EAGLE-3 decoding by setting --speculative-algorithm EAGLE3 and choosing an appropriate model.

python3 -m sglang.launch_server \
    --model meta-llama/Meta-Llama-3.1-8B-Instruct \
    --speculative-algorithm EAGLE3 \
    --speculative-draft-model-path jamesliu1/sglang-EAGLE3-Llama-3.1-Instruct-8B \
    --speculative-num-steps 3 \
    --speculative-eagle-topk 4 \
    --speculative-num-draft-tokens 16 \
    --mem-fraction-static 0.7 \
    --cuda-graph-max-bs 8 \
    --dtype float16 \
    --log-level warning

Send a request:

import openai

client = openai.Client(base_url="http://127.0.0.1:30000/v1", api_key="None")

response = client.chat.completions.create(
    model="meta-llama/Meta-Llama-3.1-8B-Instruct",
    messages=[
        {"role": "user", "content": "List 3 countries and their capitals."},
    ],
    temperature=0,
    max_tokens=64,
)

print(response.choices[0].message.content)

Multi Token Prediction

We support MTP (Multi-Token Prediction) in SGLang by using speculative decoding. We use XiaomiMiMo/MiMo-7B-RL as an example here (for DeepSeek MTP usage, refer to deepseek_v32 doc).

python3 -m sglang.launch_server \
    --model XiaomiMiMo/MiMo-7B-RL \
    --host 0.0.0.0 \
    --trust-remote-code \
    --speculative-algorithm EAGLE \
    --speculative-num-steps 1 \
    --speculative-eagle-topk 1 \
    --speculative-num-draft-tokens 2 \
    --mem-fraction-static 0.7 \
    --cuda-graph-max-bs 8 \
    --log-level warning

Send a request:

import requests

url = "http://localhost:30000/v1/chat/completions"

data = {
    "model": "XiaomiMiMo/MiMo-7B-RL",
    "messages": [{"role": "user", "content": "What is the capital of France?"}],
}

response = requests.post(url, json=data)
print(response.json())

Standalone Speculative Decoding (Small Draft Model)

Besides EAGLE/MTP, SGLang also supports token-level speculative decoding using a smaller draft model. Enable it with --speculative-algorithm STANDALONE and provide a draft model via --speculative-draft-model-path.

Relevant parameters:

Parameter	Description	Default
`--speculative-draft-model-path`	Draft model weights (smaller than the target model).	`None`
`--speculative-num-steps`	Draft depth (how many steps the draft model runs autoregressively).	`3` (auto default for STANDALONE)
`--speculative-eagle-topk`	Branching factor (token candidates per step).	`1` (auto default for STANDALONE)
`--speculative-num-draft-tokens`	Verification capacity.	`4` (auto default for STANDALONE)
`--speculative-draft-model-quantization`	Quantization for the draft model. Use `"unquant"` to disable quantization on the draft even when the target is quantized.	Same as target

Note: Standalone speculative decoding currently does not support --enable-dp-attention.

python3 -m sglang.launch_server \
    --model Qwen/Qwen2.5-7B-Instruct \
    --speculative-algorithm STANDALONE \
    --speculative-draft-model-path Qwen/Qwen2.5-1.5B-Instruct \
    --speculative-num-steps 4 \
    --speculative-eagle-topk 2 \
    --speculative-num-draft-tokens 7 \
    --mem-fraction-static 0.7 \
    --cuda-graph-max-bs 8 \
    --log-level warning

Send a request:

import openai

client = openai.Client(base_url="http://127.0.0.1:30000/v1", api_key="None")

response = client.chat.completions.create(
    model="Qwen/Qwen2.5-7B-Instruct",
    messages=[
        {"role": "user", "content": "List 3 countries and their capitals."},
    ],
    temperature=0,
    max_tokens=64,
)

print(response.choices[0].message.content)

Speculative Decoding V2 (Overlap Scheduler)

SGLang provides an experimental Speculative Decoding V2 implementation that enables an overlap scheduler and uses V2 speculative workers (e.g. StandaloneWorkerV2, EAGLEWorkerV2).

To enable it, set the environment variable:

SGLANG_ENABLE_SPEC_V2=True

Notes:

SpecV2 currently only supports --speculative-eagle-topk 1. When SpecV2 is enabled, set --speculative-eagle-topk 1 explicitly.
If you explicitly set --speculative-eagle-topk > 1, the server will error.
If you omit --speculative-eagle-topk, auto-tuning may pick topk > 1 for some models (e.g. Llama). This is incompatible with SpecV2 and may not always trigger an immediate config error, so set --speculative-eagle-topk 1 explicitly.
This applies to EAGLE, EAGLE3, and STANDALONE.

SGLANG_ENABLE_SPEC_V2=True python3 -m sglang.launch_server \
    --model Qwen/Qwen2.5-7B-Instruct \
    --speculative-algorithm STANDALONE \
    --speculative-draft-model-path Qwen/Qwen2.5-1.5B-Instruct \
    --speculative-num-steps 4 \
    --speculative-eagle-topk 1 \
    --speculative-num-draft-tokens 5 \
    --mem-fraction-static 0.7 \
    --cuda-graph-max-bs 8 \
    --log-level warning

Send a request:

import openai

client = openai.Client(base_url="http://127.0.0.1:30000/v1", api_key="None")

response = client.chat.completions.create(
    model="Qwen/Qwen2.5-7B-Instruct",
    messages=[
        {"role": "user", "content": "List 3 countries and their capitals."},
    ],
    temperature=0,
    max_tokens=64,
)

print(response.choices[0].message.content)

Ngram Speculative Decoding

SGLang also supports ngram-based speculative decoding (no separate draft model). It retrieves draft tokens from an ngram cache built from previously generated tokens, and then verifies them with the target model.

Enable it with:

--speculative-algorithm NGRAM

Ngram-specific parameters

Parameter	Description	Default
`--speculative-num-draft-tokens`	Number of draft tokens verified per step. If omitted, defaults to `--speculative-ngram-max-match-window-size`.	`12` (with default ngram settings)
`--speculative-ngram-min-match-window-size`	Minimum matching window size.	`1`
`--speculative-ngram-max-match-window-size`	Maximum matching window size.	`12`
`--speculative-ngram-min-bfs-breadth`	Minimum BFS breadth.	`1`
`--speculative-ngram-max-bfs-breadth`	Maximum BFS breadth.	`10`
`--speculative-ngram-match-type`	Match type: `"BFS"` or `"PROB"`.	`"BFS"`
`--speculative-ngram-branch-length`	How many recent tokens to insert into the cache.	`18`
`--speculative-ngram-capacity`	Cache capacity (number of entries).	`10,000,000`

Notes:

Ngram speculative decoding only supports CUDA.
It currently does not support --enable-dp-attention.
It disables the overlap scheduler and mixed chunked prefill.
If --speculative-ngram-max-bfs-breadth > 1 (thus speculative_eagle_topk > 1) and page_size > 1, use --attention-backend flashinfer; otherwise the server will error.
Optional: set SGLANG_NGRAM_FORCE_GREEDY_VERIFY=True to force greedy verification.

python3 -m sglang.launch_server \
    --model Qwen/Qwen2.5-7B-Instruct \
    --speculative-algorithm NGRAM \
    --speculative-num-draft-tokens 16 \
    --speculative-ngram-max-match-window-size 12 \
    --speculative-ngram-max-bfs-breadth 10 \
    --mem-fraction-static 0.7 \
    --cuda-graph-max-bs 8 \
    --log-level warning

Send a request:

import openai

client = openai.Client(base_url="http://127.0.0.1:30000/v1", api_key="None")

response = client.chat.completions.create(
    model="Qwen/Qwen2.5-7B-Instruct",
    messages=[
        {"role": "user", "content": "List 3 countries and their capitals."},
    ],
    temperature=0,
    max_tokens=64,
)

print(response.choices[0].message.content)

Full Parameter Reference

Below is a comprehensive list of all speculative decoding parameters available in SGLang:

Core parameters

Parameter	Type	Default	Description
`--speculative-algorithm`	`str`	`None`	Algorithm to use: `EAGLE`, `EAGLE3`, `STANDALONE`, `NGRAM`, `NEXTN` (alias of `EAGLE`)
`--speculative-draft-model-path`	`str`	`None`	Path to the draft model weights
`--speculative-draft-model-revision`	`str`	`None`	Specific revision/commit of the draft model (`"main"` is auto-used when draft path is set and revision is omitted)
`--speculative-draft-load-format`	`str`	`None`	Load format for draft model weights
`--speculative-num-steps`	`int`	`None` (auto-chosen when omitted)	Autoregressive drafting depth
`--speculative-eagle-topk`	`int`	`None` (auto-chosen when omitted)	Branching factor per drafting step
`--speculative-num-draft-tokens`	`int`	`None` (auto-chosen when omitted)	Maximum number of draft tokens for verification
`--speculative-accept-threshold-single`	`float`	`1.0`	Single-token acceptance threshold
`--speculative-accept-threshold-acc`	`float`	`1.0`	Accumulated acceptance threshold
`--speculative-token-map`	`str`	`None`	Path to FR-Spec high-frequency token map
`--speculative-attention-mode`	`str`	`"prefill"`	Attention mode for speculative operations (`"prefill"` or `"decode"`)
`--speculative-draft-attention-backend`	`str`	`None`	Override attention backend for the draft model
`--speculative-moe-runner-backend`	`str`	`None`	MoE runner backend for the draft model
`--speculative-moe-a2a-backend`	`str`	`None`	MoE all-to-all backend for the draft model
`--speculative-draft-model-quantization`	`str`	Same as target	Quantization for the draft model (`"unquant"` to disable)

Ngram-specific parameters

Parameter	Type	Default	Description
`--speculative-ngram-min-match-window-size`	`int`	`1`	Minimum ngram matching window
`--speculative-ngram-max-match-window-size`	`int`	`12`	Maximum ngram matching window
`--speculative-ngram-min-bfs-breadth`	`int`	`1`	Minimum BFS breadth
`--speculative-ngram-max-bfs-breadth`	`int`	`10`	Maximum BFS breadth
`--speculative-ngram-match-type`	`str`	`"BFS"`	Match type: `"BFS"` or `"PROB"`
`--speculative-ngram-branch-length`	`int`	`18`	Recent tokens to insert into cache
`--speculative-ngram-capacity`	`int`	`10,000,000`	Cache capacity

Environment variables

Variable	Default	Description
`SGLANG_ENABLE_SPEC_V2`	`False`	Enable Speculative Decoding V2 (overlap scheduler)
`SGLANG_NGRAM_FORCE_GREEDY_VERIFY`	`False`	Force greedy verification for ngram decoding

Other related flags

Parameter	Description
`--enable-multi-layer-eagle`	Enable multi-layer EAGLE (auto-enabled for MiMoV2 and Step3p5 models)
`--enable-torch-compile`	Enable `torch.compile` for kernel-level optimizations
`--torch-compile-max-bs`	Maximum batch size for `torch.compile`

OOM Troubleshooting

Out of Memory (OOM)? Speculative decoding may increase GPU memory usage because the draft tree, CUDA graphs, and verification-related buffers consume additional VRAM. If you encounter OOM errors, try the following adjustments.

Step 1: Lower static memory fraction (most effective)

--mem-fraction-static 0.5   # when omitted, this value is auto-computed

--mem-fraction-static controls the memory budget for model weights + KV cache pool.
Lowering it directly increases dynamic headroom for activations and CUDA graph buffers.
If omitted, SGLang auto-estimates this value from other settings, and those auto settings can still be too aggressive for some workloads.

Step 2: Reduce CUDA graph batch size

# Fewer CUDA graph captures = less memory reserved
--cuda-graph-max-bs 4   # or even 2 for tight memory situations

If omitted, --cuda-graph-max-bs is auto-selected based on GPU memory and TP size, and can be much larger on high-memory GPUs.

Step 3: Reduce draft tree size

These three parameters directly control how much memory the draft tree consumes:

# Before (aggressive, high memory)
--speculative-num-steps 5 --speculative-eagle-topk 8 --speculative-num-draft-tokens 64

# After (conservative, lower memory)
--speculative-num-steps 3 --speculative-eagle-topk 1 --speculative-num-draft-tokens 4

Step 4: Limit concurrent requests

# Fewer concurrent requests lowers in-flight load and can reduce OOM risk
--max-running-requests 4

Quick OOM recovery recipe

If you're hitting OOM and just want something that works, start with this minimal configuration and scale up:

python3 -m sglang.launch_server \
    --model <your-model> \
    --speculative-algorithm EAGLE \
    --speculative-draft-model-path <your-draft-model> \
    --speculative-num-steps 3 \
    --speculative-eagle-topk 1 \
    --speculative-num-draft-tokens 4 \
    --cuda-graph-max-bs 2 \
    --mem-fraction-static 0.5 \
    --max-running-requests 4 \
    --log-level warning

Then gradually increase --speculative-num-draft-tokens, --speculative-eagle-topk, and --cuda-graph-max-bs. Increase --mem-fraction-static last, only after the run is stable.

References

EAGLE process is as follows:

Within EAGLE the draft model predicts the next feature vector, i.e. the last hidden state of the original LLM, using the feature sequence $(f_1, ..., f_k)$ and the token sequence $(t_2, ..., t_{k+1})$.
The next token is then sampled from $p_{k+2}=\text{LMHead}(f_{k+1})$. Afterwards, the two sequences are extended in a tree style—branching out multiple potential continuations, with the branching factor per step controlled by the speculative_eagle_topk parameter—to ensure a more coherent connection of context, and are given as input again.
In SGLang's EAGLE-2 implementation, the draft tree is expanded for the configured steps and then reranked to select the top speculative_num_draft_tokens final nodes as draft tokens.
EAGLE-3 removes the feature prediction objective, incorporates low and mid-layer features, and is trained in an on-policy manner.

This enhances drafting accuracy by operating on features instead of tokens for more regular inputs and by additionally passing tokens from the next timestep to reduce sampling randomness. For more details, see the EAGLE-2 and EAGLE-3 papers.

For guidance on how to train your own EAGLE model please see the EAGLE repo. For EAGLE-3 training specifically, check out SpecForge, the SGLang team's training framework designed for EAGLE-3 speculative decoding models with seamless porting to SGLang serving. See the SpecForge documentation and blog post for details.

Speculative Decoding

Summary

Jump to sections

Quick guidance

Method comparison (mini table)

Performance Highlights

EAGLE Decoding

EAGLE-2 Decoding

EAGLE-2 Decoding with torch.compile

EAGLE-2 Decoding via Frequency-Ranked Speculative Sampling

EAGLE-3 Decoding

Multi Token Prediction

Standalone Speculative Decoding (Small Draft Model)

Speculative Decoding V2 (Overlap Scheduler)

Ngram Speculative Decoding

Ngram-specific parameters

Full Parameter Reference

Core parameters

Ngram-specific parameters

Environment variables

Other related flags

OOM Troubleshooting

Step 1: Lower static memory fraction (most effective)

Step 2: Reduce CUDA graph batch size

Step 3: Reduce draft tree size

Step 4: Limit concurrent requests

Quick OOM recovery recipe

References

EAGLE-2 Decoding with `torch.compile`