Home
About
Blog
Products
Forum
Support
Contact
Sunbelt Computer Software
PL/B Language Development and Support
Home
About
Blog
Products
Forum
Support
Contact
semblend/semblend/__init__.py at main · WorldFlowAI/semblend · GitHub
Skip to content
Navigation Menu
Sign in
Appearance settings
Platform
AI CODE CREATION
GitHub Copilot
Write better code with AI
GitHub Copilot app
Direct agents from issue to merge
MCP Registry
Integrate external tools
DEVELOPER WORKFLOWS
Actions
Automate any workflow
Codespaces
Instant dev environments
Issues
Plan and track work
Code Review
Manage code changes
Code Quality
Enforce quality at merge
APPLICATION SECURITY
GitHub Advanced Security
Find and fix vulnerabilities
Code security
Secure your code as you build
Secret protection
Stop leaks before they start
EXPLORE
Why GitHub
Documentation
Blog
Changelog
Marketplace
View all features
Solutions
BY COMPANY SIZE
Enterprises
Small and medium teams
Startups
Nonprofits
BY USE CASE
App Modernization
DevSecOps
DevOps
CI/CD
View all use cases
BY INDUSTRY
Healthcare
Financial services
Manufacturing
Government
View all industries
View all solutions
Resources
EXPLORE BY TOPIC
AI
Software Development
DevOps
Security
View all topics
EXPLORE BY TYPE
Customer stories
Events & webinars
Ebooks & reports
Business insights
GitHub Skills
SUPPORT & SERVICES
Documentation
Customer support
Community forum
Trust center
Partners
View all resources
Open Source
COMMUNITY
GitHub Sponsors
Fund open source developers
PROGRAMS
Security Lab
Maintainer Community
GitHub Stars
Archive Program
REPOSITORIES
Topics
Trending
Collections
Enterprise
ENTERPRISE SOLUTIONS
Enterprise platform
AI-powered developer platform
AVAILABLE ADD-ONS
GitHub Advanced Security
Enterprise-grade security features
Copilot for Business
Enterprise-grade AI features
Premium Support
Enterprise-grade 24/7 support
Pricing
Search
/
Sign in
Sign up
Appearance settings
You signed in with another tab or window.
Reload
to refresh your session.
You signed out in another tab or window.
Reload
to refresh your session.
You switched accounts on another tab or window.
Reload
to refresh your session.
Dismiss alert
{{ message }}
Uh oh!
There was an error while loading.
Please reload this page
.
WorldFlowAI
/
semblend
Public
Notifications
You must be signed in to change notification settings
Fork
2
Star
4
Code
Issues
1
Pull requests
0
Actions
Projects
Security and quality
0
Insights
Additional navigation options
Code
Issues
Pull requests
Actions
Projects
Security and quality
Insights
Files
Expand file tree
main
Breadcrumbs
semblend
/
semblend
/
__init__.py
Copy path
Blame
More file actions
Blame
More file actions
Latest commit
History
History
History
136 lines (126 loc) · 3.79 KB
main
Breadcrumbs
semblend
/
semblend
/
__init__.py
Copy path
Top
File metadata and controls
Code
Blame
136 lines (126 loc) · 3.79 KB
Raw
Copy raw file
Download raw file
Open symbols panel
Edit and raw actions
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
"""SemBlend - semantic donor discovery for engine-native KV caches.
Extends exact-prefix caching in inference engines with semantic donor discovery:
finds safe donors from semantically similar prior requests and plans reuse of
engine-owned KV state, eliminating redundant prefill computation on suitable
long-context workloads.
Quick start:
pip install semblend
# vLLM compatibility path (dynamic connector loading):
vllm serve model_name
\\
--kv-transfer-config '{"kv_connector": "SemBlendConnectorV1",
"kv_connector_module_path": "semblend.integration.vllm.connector_v1",
"kv_role": "kv_both"}'
# SGLang integration (HiCache storage backend):
python -m sglang.launch_server --model-path model_name
\\
--enable-hierarchical-cache
\\
--hicache-storage-backend dynamic
\\
--hicache-storage-backend-extra-config
\\
'{"module_path":"semblend.integration.sglang.hicache_backend",
"class_name":"SemBlendHiCacheStorage"}'
Environment variables:
SEMBLEND_ENABLED=1 Enable semantic donor search
SEMBLEND_MIN_SIMILARITY=0.60 Minimum cosine similarity threshold
SEMBLEND_EMBEDDER=minilm Embedder type (minilm, jaccard, onnx_gpu)
SEMBLEND_FUZZY_CHUNKS=1 Enable fuzzy chunk matching
"""
from
__future__
import
annotations
from
importlib
.
metadata
import
PackageNotFoundError
from
importlib
.
metadata
import
version
as
_dist_version
try
:
# Single source of truth is the installed distribution; a hand-kept
# literal here shipped stale in 0.3.17 and tripped engine version gates.
__version__
=
_dist_version
(
"semblend"
)
except
PackageNotFoundError
:
# running from a source tree without install
__version__
=
"0.0.0+source"
# Re-export core public API from semblend_core
from
semblend_core
.
alignment
import
(
AlignmentResult
,
SlotAction
,
SlotActionType
,
compute_alignment
,
compute_batch_alignment
,
compute_chunk_alignment
,
compute_exact_run_alignment
,
estimate_reuse_ratio
,
)
from
semblend_core
.
backend
import
SemBlendBackend
from
semblend_core
.
bathtub
import
(
BathtubPreset
,
LayerDeviation
,
compute_layer_deviations
,
get_preset
,
sigma
,
)
from
semblend_core
.
donor_store
import
DonorMatch
,
DonorNode
,
DonorStore
from
semblend_core
.
embedder
import
(
EmbedderType
,
JaccardEmbedder
,
MiniLMEmbedder
,
OnnxGpuEmbedder
,
create_embedder
,
)
from
semblend_core
.
partial_attention
import
(
AttentionMode
,
PartialAttentionPlan
,
build_attention_plan
,
compute_attention_mask
,
compute_donor_kv_indices
,
)
from
semblend_core
.
pipeline
import
(
PipelineResult
,
PipelineTimings
,
PositionMapping
,
SemBlendPipeline
,
)
from
semblend_core
.
simhash
import
(
bulk_hamming_distance
,
compute_simhash
,
hamming_distance
,
is_plausible_donor
,
)
__all__
=
[
"__version__"
,
# Pipeline
"SemBlendPipeline"
,
"PipelineResult"
,
"PipelineTimings"
,
"PositionMapping"
,
# Backend ABC
"SemBlendBackend"
,
# Alignment
"AlignmentResult"
,
"SlotAction"
,
"SlotActionType"
,
"compute_alignment"
,
"compute_batch_alignment"
,
"compute_chunk_alignment"
,
"compute_exact_run_alignment"
,
"estimate_reuse_ratio"
,
# Donor store
"DonorMatch"
,
"DonorNode"
,
"DonorStore"
,
# Embedder
"EmbedderType"
,
"JaccardEmbedder"
,
"MiniLMEmbedder"
,
"OnnxGpuEmbedder"
,
"create_embedder"
,
# Bathtub
"BathtubPreset"
,
"LayerDeviation"
,
"compute_layer_deviations"
,
"get_preset"
,
"sigma"
,
# Partial attention
"AttentionMode"
,
"PartialAttentionPlan"
,
"build_attention_plan"
,
"compute_attention_mask"
,
"compute_donor_kv_indices"
,
# SimHash
"bulk_hamming_distance"
,
"compute_simhash"
,
"hamming_distance"
,
"is_plausible_donor"
,
]
You can’t perform that action at this time.